Yeah, this is issue that Scirra doesn't add plugins to access Node-webkit API

I really can't imagine why you'd want to directly access the Node-Webkit API through a graphical interface.

What would you even gain from this?

Other than changing the icon (which doesn't really require a new plugin but rather the exporter to add more user input to the package.json than it does right now), I can't think of a simple reason.

Having access to a specific API isn't magically transforming all your code and what little you might want to add can already be done by using the browser object.

tanoshimi

Nice.

That's what I did for my last two presentations, as well.

Works like a charm

It's definitely possible. It also wouldn't work well as a first party plugin, since there's more than one CLI (there's not just the Windows CMD but also Linux/Unix shells), which would make the whole implementation a bit messy (IMHO).

Lastly, I second zenox98's notion. We shouldn't blame Ashley for every miserable little thing. He might be the creator of Construct2, but he's not the creator of Node Webkit and thus certain issues might be out of his control (like the lack of support for several third party plugins or the weird looking, chewed off alert window in Node Webkit. Blaming Ashley for the lack of customization options for Node Webkit won't help either)

Well. You could add a textbox to the layout and, instead of putting down an URL, tell "Load Image from URL" to load the image from the URL entered by the user inside that textbox. In short: Instead of an URL, you would simply write down "Textbox.Text".

However, a better way to load sprites may be the drag and drop behaviour introduced with this plugin:

scirra.com/forum/plugin-dragndrop-zone_topic58578.html

I was facing a similar problem (I needed this for several sprite based editors inside my point and click adventure engine) when lemo was kind enough to tell me about it.

sogen

MySQL is for databases. It wouldn't make sense to query a single image through a database (except if you wanted to make an MMORPG/graphical MUD and save several characters' outfits on a server). AJAX is what you're looking for, which essentially is what the Load Image from URL object is mimicing.

Those links might also be quite helpful:

iquilezles.org/apps/shadertoy

html5rocks.com/en/tutorials/webgl/shaders

chromeexperiments.com/webgl

webglplayground.net/gallery

Your best source for learning about shaders in general are still chipmakers, though.

NVIDIA's developer zone is unmatched, IMHO.

Is there some sort of UK legal entity I can apply with, in order to get my project hosted on Kickstarter? Preferably a reliable one that isn't keeping anywhere from 20-50% of the Kickstarter pledges as "processing fee".   

I'm not too bullish about a possible Kickstarter Germany (if it ever happens).

>

> Using my language flags inside the title tag (i. e. <div class="lang:english">Home</div><div class="lang:german">Startseite</div>) further decreased my rating, unfortunately.

>

> Which bares the question: Why is there no legit way to have a multilingual title tag yet?

Why don't you use PHP, ASP.NET or whatever and detect the browser language / user location if you have a website where you need to care about multilingualism?

1. PHP is an inefficient, unelegant POS. No way in hell I'm ever using it for any bigger homepage project.

2. My homepage uses Google Blogger, so there's no way to use ASP.Net. There might be a way to integrate with my favourite choice, Python, through the use of Google App Engine. I'm still working it out, though. Implementing a Python based ecommerce platform will be a higher priority anyways.

3. For what I'm doing, Javascript is way faster and more efficient. No need to reload a page just to switch languages. No people at an internet caf?/on their holidays getting mad at me because I chose to detect a language they might not even be fluent at. No need to keep track of multiple pages or an oversized sitemap.

Construct2 should definitely be able to handle it.

As a WebOS user first and Android user second, I haven't yet felt the need to use CocoonJS/DirectCanvas. But from what I can tell, Phonegap adds very little overhead.

An HTML5 based ebook/children's book should definitely be possible and has, in fact, been done multiple times already (Kindle Format 8).

As for native toolsets. I'd recommend either Gideros (there's a children's book about a small blue elephant featured right on their homepage) or Haxe NME.

Gideros already comes with it's own IDE and Haxe NME supports plenty of other exciting platforms from the get-go.

NVIDIA's GPU Gems is a great place to get started.

Most of their shaders are HLSL though, so you'll have to port them to GLSL first. More complex shaders also may not work with WebGL (which is OpenGL ES in a bubble, basically)

Whiteclaws

I'm doing this so I can complete my game. Do you not understand what kickstarter/Indiegogo is for?

He has a point though.

At the very least add some non-programmer art.

Seeing as how your game and the Joomla page are hosted by different hosts, I'd refer you to the use of iframes.

It's good enough for cross-domain content, though should be avoided if the iframe is pointing to the same domain it's on.

AFAIK, the only way to achieve this is by using the custom pathfinding behaviour (thus abandoning solid entities in favour of explicitly defined obstacles).

That way, you can also clear the obstacle map in case of a specifically scripted event (quite handy if you have cutscenes defying the laws of physics or simply want your guard run past all obstacles, looking for a spot to hide, in case the player missed the shot).