|
I think this smells of spam is what I think.
|
|
|
|
|
Spam? Well, not really- I am not advertising nor anything. Unreal Engine a well-known game development engine that is considered as the most powerful real time rendering tool.
So back to the subject, what does a software engineer/developer think of visual scripting environment? 
|
|
|
|
|
 I do not think so. If you are even nowhere near game development, you must have heard from the unreal engine.
(Otherwise check my sig link  Now that was real spam  )
|
|
|
|
|
Rage wrote: If you are even nowhere near game development playing
It's enough! Apart from Unreal most games are based on its engine (and knowing the engine a game is based on gives a lot of insight on the mechanics and if it's worthy to be bought or not).
GCS d-- s-/++ a- C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- ++>+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
|
Pros: Allows rapid experimentation/feedback loop
Cons: Always something you want to do that the designers didn't think of
|
|
|
|
|
Well, that's relatively correct. however, such advanced visual scripting environments give remarkable tools to make the system more extensible. For example, designers have already implemented an approach that makes the users able to write their own nodes/(blocks of code).
|
|
|
|
|
Precisely so - the scripting / code block part is a work around for scenarios where the visual designer does not match to the problem space.
|
|
|
|
|
You've got a very good point!
Usually, people tend to download, purchase and create plugins to fill the void, which is- of course, another workaround to use some tools that the designers FORGOT about.
|
|
|
|
|
In my experience a good approach is to have the designer generate code in the form of partial classes which the developer can override/augment with their own crafted code in their side of the partial classes.
Many ETL designers and things like the EF designer work in this way.
|
|
|
|
|
Yeah, that's always the way. Same as in SSIS (and probably all ETL systems). And also in a rule-based system I had to use a few years ago.
All these graphical drag-and-drop purveyors are quick to say "never write another line of code". But then when you do have to write some code it's ten times more difficult.
|
|
|
|
|
The concept has been tried multiple times, and most of us are still using a text-editor. That's because a text-editor is the superiour way to quickly edit a text, which still communicates more information in a few lines than most visualizations communicate in pages.
It is a step in the wrong direction 
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
Hear! Hear!
When writing must be done it can't be done graphically. Graphical systems impose too many limitations.
|
|
|
|
|
Do Ctrl-X, Ctrl-End, Ctrl-V, Ctrl-A, Ctrl-SHIFT-U, Ctrl-KC.
Quick edit in text, now show me your productivity-increase with a visual editor and perform the same tasks 
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
Was that intended for me?
|
|
|
|
|
No, but I don't see anyone else either; it is one of those limitations that really bugged me when on a graphical system.
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
Message Closed
modified 19-Feb-18 12:33pm.
|
|
|
|
|
I'll bet you a banana that we'll still prefer text over visuals in 20 years
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
Particularly as I could be blind by then. 
|
|
|
|
|
I have found that such tools address only the simplest tasks anyway -- the developers went after only the "low hanging fruit", when you don't need a tool to do those things to begin with!
We need tools that help with the difficult tasks. Instead they take the easy things and make them simple -- "hey, we saved 90%" -- while not helping with the difficult tasks and actually making the difficult even more difficult to do in that framework/tool.
And this still smells spammy to me.
|
|
|
|
|
PIEBALDconsult wrote: "hey, we saved 90%"
Hence the old planning adage "The first 90% takes 90% of the time, and the last 10% takes the other 90% of the time"
|
|
|
|
|
Quote: I have found that such tools address only the simplest tasks anyway
Basically, that's only correct in some specific cases. However, when talking about products such as the blueprints system of unreal engine, things become totally different.
In this VIDEO, you can see how one person made an advanced video game that is now seen as the most awaited RPG game in 2018- not mention that it is, seriously, free of any single line of code. It is pure blueprints.
Quote: We need tools that help with the difficult tasks. Instead they take the easy things and make them simple -- "hey, we saved 90%" -- while not helping with the difficult tasks and actually making the difficult even more difficult to do in that framework/tool.
You are perfectly correct. We need sophisticated tools that help us with the difficult tasks. That's why tools such as visual scripting environments do exist.
The project that I mentioned above, which is 100% blueprints( no single line of codes) has been adopted by a game development studio, and they continued working on the project using blueprints only.
Here is the game after the adoption (still, no line of code):
Lost Soul Aside PSX2017 Demo Full Gameplay - YouTube[^]
|
|
|
|
|
|
It makes me think of people that builds homepages in Word.
The code created is utterly horrible.
|
|
|
|
|
In a lot of ways using a visual scripting editor is not too far removed from high level scripting and programming languages.
It's really just another abstraction and most of the decent ones will behave in the following manner:
(1) I start with the visual designer.
(2) I then manually refactor or extend the code created by the designer.
(3) I use visual designer again to add a high level function/module.
(4) I go back to manually extending the code created by the visual designer at (3).
If at point (4) I discover that the tool has refactored all my changes from (2) then it's a fail with regards to the visual designer as far as I am concerned.
A decent visual designer will not refactor my code at (3).
“That which can be asserted without evidence, can be dismissed without evidence.”
― Christopher Hitchens
|
|
|
|
|
Like Windows Workflow Foundation (WF) [^].
At first I was contrary because there is ALWAYS the need or reason to do things flexibly... experience taught me that it's still fundamental in about 30 to 40% of the cases and that having an easy, graphical, standard even if not flexible tool to quickly prototype functionalities and sequences can improve the productivity a lot.
Think of the AI patterns of videogames, or massive online games with transactions and interactions between weapons, AI, effects, powers, passive abilities, resources and currencies all spread between millions of clients and a plethora of servers located across the world. A single bugfix, done only textually, could take weeks - if given a scalable way to prototype most actions with a 1:1 relationship between prototype and code it could take HOURS.
I believe Digital Extremes uses an approach like this, as they are able to churn out an update a week and 3-4 hotfixes in hours - and most of their talk is about mechanics of the game.
GCS d-- s-/++ a- C++++ U+++ P- L+@ E-- W++ N+ o+ K- w+++ O? M-- V? PS+ PE- Y+ PGP t+ 5? X R+++ tv-- b+(+++) DI+++ D++ G e++ h--- ++>+++ y+++* Weapons extension: ma- k++ F+2 X
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
