Thread: Why you should develop your plugin for nGhost3

Originally Posted by matt11601

@Kev000
Many people continue to use Windows front ends because the best front ends RUN on Windows. In the future, I think it would help users to see a chart/comparison of what nGhost can do compared to, say, Ride Runner or Centrafuse. Once users see a "feature" comparison, I think the switch will be easier. With this increased interest I'm hoping there will be more exposure, user created plug-ins, etc.

What do you think?

I agree, but also disagree. While I have used different flavors of linux in the past - I know Windows inside and out. The amount of time it would take me to get a good setup using linux that I was happy and familiar with - Outweighs the amount of time i'd like to put into it unfortunately.

When something breaks or doesn't work in windows - I can fix it in minutes. When somethings not working right, won't install correct or gives me issues - takes me hours/days to figure out.

I understand than linux is much more efficient than windows, and everything is endless and open ended - But I just don't have the time or patience to figure it all out. I ran Ubuntu ("arguably" the most newb friendly version of linux) for a solid month before I switched back. I still don't think i'd be able to install an application without some help.

Originally Posted by kev000

nGhost3 probably won't run well on windows. This is because while other frontends hide/mask the shell, nGhost IS the shell and windows, well, doesn't play nice with others. Someone with a lot more windows knowledge than me could probably explain why killing the windows shell (iirc explorer.exe) is a bad idea.

Yes, explorer.exe is the Windows shell. That's configurable though, you can have it start nghost3.exe instead of explorer.exe if you want. Explorer would no longer start up, unless of course you launch it yourself, and killing nghost.exe would then have the same side-effects as killing explorer.exe (haven't messed with this in a long time, but if I recall correctly, Windows will automatically launch a new shell process if the first one dies).

Of course that assumes that you first create yourself an nghost.exe for Windows to use as the shell. But if you do that, then you can replace explorer.exe with your own shell pretty easily.

I'm guessing that the biggest three factors that guide which front end a developer writes a plug in for are:
1) Which front-ends have the most users? (if they're in it for the satisfaction that comes from having people use their code)
2) Which front-end does the individual developer like best (if they're in it for the satisfaction of using something that they wrote, or something that has a feature that they particularly want)
3) Which front-ends have the most developer-friendly API and documentation? (if speed of development is a big motivator)

Any new front end is going to be in a tricky place with regard to (1) because users tend to want the front ends that have their favorite plugsins. Classic chicken-or-egg problem. But (2) and (3) and largely under your control, and I think you're doing the right thing by focusing on the core features (nav, music, phone) and creating APIs for both native code and C#.

I'm not using nGhost but I am keeping an eye on it and I look forward to watching it evolve.

Originally Posted by matt11601

@Kev000
It looks like nGhost is gaining exposure and interest from mp3car forum users. I think later versions of nGhost (perhaps version 3?) will inspire users to make the jump to linux and ultimately write plug-ins (hence the purpose of this thread).

I do have a suggestion for a future thread. Many people continue to use Windows front ends because the best front ends RUN on Windows. In the future, I think it would help users to see a chart/comparison of what nGhost can do compared to, say, Ride Runner or Centrafuse. Once users see a "feature" comparison, I think the switch will be easier. With this increased interest I'm hoping there will be more exposure, user created plug-ins, etc.

What do you think?

Another thing that would help is a video walking through the features of nGhost. Start by picking a song or playlist from the music library, placing a call / taking a call, displaying nav, etc, then go into more detail about each feature. Swapping operating systems is a non-trivial undertaking, people are going to need to be convinced that it's worth investing the time.

A short video walkthrough would probably help. At least have screenshots. But I think something showing hands-on (literally) usage would help even more than screenshots.

I just reworked the animation sublayer in the nghost3 libraries to allow more advanced effects. This is a short demo featuring the "dialer" example and the new click effect:

[media]http://www.youtube.com/watch?v=nmKej6GJsTU[/media]

Originally Posted by kev000

I just reworked the animation sublayer in the nghost3 libraries to allow more advanced effects. This is a short demo featuring the "dialer" example and the new click effect:

[media]http://www.youtube.com/watch?v=nmKej6GJsTU[/media]

Very nice demo. I think the pop-up effect upon selection is more than eye candy because it will give the user visual confirmation of key presses if they're using a touch screen. Will users have the option to adjust the amount of "pop-up time?" Are other effects planned such as the numbers spinning or fading/re-appearing upon press?

Originally Posted by matt11601

Very nice demo. I think the pop-up effect upon selection is more than eye candy because it will give the user visual confirmation of key presses if they're using a touch screen. Will users have the option to adjust the amount of "pop-up time?" Are other effects planned such as the numbers spinning or fading/re-appearing upon press?

I want to make things like effect, mode, and duration configurable via the theme. That won't come until the theme engine is locked down though. Mode really defines how the effect looks. This is an elastic-out mode and that's why you see the bouncing.

I've got animation "presets" already created:
-fade to 20%
-fade-out
-fade-in
-grow to 50% (default button behavior)
-spin 360 on the x-axis
-spin 360 on the y-axis.

Of course, you can also mix multiple animations together like what you see in the keyboard demo on the first post.

Here's another demo, this time in the form of an actual functioning application. Indeed the first "real" application to use the new ng3 framework:

icepanel3

[media]http://www.youtube.com/watch?v=LYWfyUEBoQQ[/media]

what's new in icepanel3 over icepanel2?

-Better integration with window manager
-Integration with compiz's scale plugin (the window selector feature)
-Integrates with ng2 through ng2's dbus plugin (will do the same with ng3).
-Replaces the window title bar functionality (notably the "X" close button).

New concepts:

-A touch-friendly dock allows you to use any application without having to Embed-The-World (TM) in a frontend. It allows you to maintain persistent controls across all windows, and maintains the same look/feel. With proper integration, you won't even know that your apps aren't embedded and are running in their own window space...
-The dock/panel replaces the window boarder/title bar. This enables the maximum screen area to be dedicated to actual content.
-Volume: click = mute/un-mute, drag adjusts the volume up/down. While this is only a concept in the video. I hope to work out the details soon so it actually works.

complete mobile but only runs on linux and macs? I think you'd need windows to get the most interest.

Also, first thing I noticed when looking at the first post (demo of keyboard) is that the keyboard changes the numbers to their "shift" alternative (!"/$%?&*())when hitting caps lock. That's not how a keyboard works, you should be able to type in caps and still have the numbers above the keypad. This is very NOT intuitive.

i have a question, does nghost require a complete restart everytime? any sort of hibernation/standby mode?