Ashley's Forum Posts

  • Folks, to be honest, at this time it looks best for us to close-source the editor as well.

    We're discussing all our options behind-the-scenes and there are a tonne of options (like taking royalties on sales of third party exporters) that are impossible with an open source editor. By having any part open-source, our options rapidly dwindle to very few limited choices. We can be an awful lot more flexible with licensing with closed source.

    Obviously this isn't aimed to rip off indie devs, I'm still in favour of a cheap or free indie/personal license. It just means that suppose Google came along and wrote an Android exporter and started selling it - we can ask for royalties (for example - this is not a definite plan). That's a neat possible revenue opportunity and it's just impossible if the editor is open source.

    As far as open-source contributions have gone in 0.x, we've had some cool patches by R0J0hound which is great and we're thankful for. However, in terms of options for making money, this is incredibly limiting, and given that we haven't exactly had hordes of developers battering down our doors demanding to write code for 0.x, it seems a logical move to go closed source.

    Again, not definite and still very much a work in progress, but it's on the cards and we're thinking it over. We don't want to rush in to any decision but I thought I'd let everyone know what we're considering. I guess it wouldn't be popular, but it makes it much more likely that we can really build something great. And the sooner we do it the better, to get ahead of anyone who forks the code already available.

    What are your thoughts? If C2 went closed source but still had a cheap or free edition, would that put you off it? How important is the open source to you?

  • The 'cannot find object expression' appears when the expression name in the AddExpression at edittime does not match the expression name at runtime. For example if your AddExpression adds "GetFoo", then you say exps["GetBar"] = function..., then it looks for "GetFoo" but can't find it.

    In your example it looks like you've deleted a crucial line for some reason. You should have something like this:

    (function()
{
	// This is the line you were missing, it creates a local 'exps' as a reference to the object expressions (this is taken from the Mouse object)
	var exps = cr.plugins.Mouse.prototype.exps;
	
	exps["X"] = function (ret) ...[/code:2myipk7s]

Without that line, you're writing the expression functions to a new global variable called 'exps', so C2 will never find the expression routine.

  • Can you post a .cap of that not working properly on the tracker?

  • In all the object methods, 'this' refers to the current instance, so just do this.now = new Date() etc.

  • Are you checking 'image' is not null? Otherwise AFAIK it'll just draw a white solid fill box. SetTexture and a Quad are all you need to get a texture on the screen though.

    (Note you should always check for NULLs - don't assume GetParamFirstInstance returns an instance - what if there are no instances?)

  • I suppose, but with good SDK documentation, the source to the exporter isn't important - and that's a fairly minor disadvantage outweighed by some solid advantages, IMO.

  • Thanks for the encouragement

    I thought opengl was badly implemented/supported on windows 7 (even more bad than on XP). I am still on XP, but that's what I've read at some point on the net. Have you ran tests, or something of the like to make sure ?

    I develop C2 on Windows 7 and haven't had any problems. It's really up to the graphics card makers to make good quality OpenGL drivers for it to work well, and it's very much in their interest to do so. Actually, so far I think it'd been working better than the old DirectX editor (no D3DX install needed), so I'm guessing you just heard rumour.

  • Anyone had any new problems with this update? Considering marking it stable and sending it out via auto-update.

  • I don't see any reason this will affect plugins. What are you worried about?

  • are you referring to the source of the exporter or the exporter itself here?

    The source code would be closed-source (proprietary). The licenses refer to how the released exporter is given away, not the source code itself, which is private to us.

    Back in time there was some legal actions threaten from CT about re-use of their coding ideas (ACE for instance)

    I don't believe there's any issue here, that all appears to have blown over. Besides, ideas aren't copyrightable - try putting IE9 and Chrome side by side.

  • I'm sorry if you're frustrated by the late notice, but it's done now. I thought it would be better to have something to show with the announcement - rather than "we won't finish 0.x, and haven't got anything else interesting to show you right now". Why would anyone believe C2 was going to be better if they couldn't see for themselves?

  • BROO, what exactly are you unhappy about? Browser makers are hardware accelerating their canvases so they'll be fast soon - check out how fast the IE9 beta is. You say 0.x is a buggy disaster yourself, doesn't that mean C2 is headed in the right direction??

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads

  • It's good idea to get some of Cons2 closed source and still be clean when it comes to CT issues.

    What CT issues? And what difference would being closed source make to that??

  • Hi all,

    Thinking more along the lines of moneymaking, I think the HTML5 exporter should be close-sourced. The code all went up on SourceForge when the preview went public, but I think it should be taken back off. Here's why:

    As you know we're still winging it as hobbyists but we like where this is going and would love it if we could go full time and make some money off it. That's good for the community too, because it means we can work fulltime on improving Construct 2, rather than the irregular patches everyone's used to.

    Corporations and businesses are a really good place to go to make money - they have lots of it and regularly license software site-wide anyway. The problem with fully open source software is it makes it impossible to sell to businesses. Even if we tried to charge, the open source licenses mean they could just get one of their employees to spend half an hour building it themselves, then they've got a copy they can legally use.

    If we close-source the HTML5 exporter, we can make sales to businesses. If we did this, there would definitely be two licenses: an indie dev/personal use license, which would either be pay-what-you-want or really cheap so you guys don't lose out (I'm siding with pay-what-you-want), and the business/commercial/site license, where we make the moolah. (Obviously the code as it is now is up there but it's still in alpha stage, and after a few months of development we'd be way ahead anyway.)

    The editor can stay open source GPL - there's no need to close source it - but without any exporters it's just an empty shell that does nothing. However, that leaves the door open to more third-party exporters that could also be free, so there's still all the main advantages of open source. So in future there could also be a totally-free edition if somebody made another free exporter. It's only the HTML5 exporter component which would be closed source.

    How does this sound? IMO, it's a win for everyone. You get a better product and we get to make some money.

    Please don't go round saying we are definitely going to go closed source and start charging money! This is just an idea under consideration right now. Let me know what you think.

  • It works like this:

    <some condition> ? <what to return if condition true> : <what to return if condition false>

    So you can put conditions directly in expressions. For example:

    + Sprite X < 100

    Set text to "Left"

    + Sprite X >= 100

    Set text to "Right"

    is equivalent to

    + Every tick

    Set text to Sprite.X < 100 ? "Left" : "Right"