Thread: An Open Source, Fast and Modular, C# Front End

Originally Posted by justchat_1

You have a good point. But not relying on other programs and designing everything from scratch are two different things. We can combine GPL code from thousands of other open source projects.

Last I heard, CF paid for a license of BlueSoliel for every copy they sell. Not to mention how much $$$ they charge for their software - free will always be appealing. I have already written the bluetooth library for another project and it was written from scratch and supports 3x the phones and features that BlueSoliel supports.

The nav engine wont be written from scratch but it will be fully integrated into the framework not just some com reference. Right now the plan is to use navit to keep with the whole 100% free 100% open source theme but the others can be integrated with lesser functionality.

But enjoy CF, enjoying paying the upgrade price to cf3 (i would guess its still at least a year away). I know you have about as many posts on their forums as the developers so I wont try to change your mind

It is true, I did pay alot, especially considering I paid for Standard, then the NAV edition. I feel that most of the cost goes to reliable customer service and ease of use and the licenses for BlueSoliel and Destinator. The CF3 upgrade price is promised to be fair, and this time Im going for Standard since I have finally gotten use to iGuidance.

Well many of you know, I hate $h!+ Deck, oopps I mean Street Deck, tried RR, but I hated all the configuring. I then tried CF, complained about not havig multi screen support, they then added it and I started on my new setup that was literally built around CF. Me and Flux had a small falling out during their work on CF3 ( they did a StreetDeck and left everyone in the wind), but they then apoligized and I realized that designing my system around CF will have me as a lifetime member, so I apoligized also for exploding and leaving a very nasty post (I have anger problems), lol.

On a side note, have you ever thought of maybe also creating the system on a diffierent OS. _Dejan_ posted a thread in the BETA area of a OS called ReactOS. Its supposed to be a clone of Windows XP/2003. Here is a link to it: http://www.reactos.org/en/index.html. Im curious to see how close to Windows this is. Maybe its faster, without all the bloat?

Originally Posted by NSFW

I suggest that one of the first features should be "launch an external app." That way, for those of us who already iGuidance and/or OBD2 software, and/or other favorite external apps, those features will be ready immediately.

But I'm kind of biased because those are my two most-used features and they're handled by standalone apps. I'd start using this new FE as soon as it can act as a shell to launch those.

Plus there will always be standalone apps that a small number of people find extremely important, that won't have enough demand to justify writing something into the front end, or that are too proprietary to integrate directly. (For example, I use closed-source ECU reflashing software for my Subaru.)

Let me add onto that, and suggest embedding application support. Strip off the title bar and embed the window into your application (winapi SetParent and SetWindowLong functions) to make them even more integrated and seamless.

Originally Posted by justchat_1

Theres a few schools of thought here...and that method and other task priority methods are in the majority. I wont go as far as to say that this software is an exception to common software design but being open source changes some things. There being (hopefully) multiple developers allows multiple items to be worked on simultaneously. The heavy reliance on a modular, plugin based architecture allows us to focus just on developing the framework and let other developers create plugins to meet user demand.

That said, planning out every little detail before we start would probably be a never ending task. My goal is to figure out the basics for each part of the framework (UI engine, Database, plugin host) and then divide up the work.
The only reason I was wondering about the rendering engine is because that is clearly the highest priority and the part I am working on first.

justchat_1, i think you missed his point of scrum. I've never used it but others at my work constantly try to push me on it You take your basic tasks and break them up into priorities. Then work only on those priorites. As new ideas come in you reorganize the priorites. Then on top of that we do sprints. Where you take the priority items and work only at that top priority for that week, month or what ever tiem frame. The of course beta and release but thats at my work.

I know you have framework development first. But you Scrim around those things first. Main priority, graphics engine, Skin layouts, pluginability, speed or wow factor .

the scrim doesn't have to be about what the front end looks like. It can be about the framework. At least its worth your time to look at the link NSFW posted.

Thats my 2 cents. i will be watching this thread(s). Good luck sounds fun.

-pRoFiT

To add another C# front end, here is mine. It uses WMP10+, skinnable with skin editor, extendable, easy install, working in my car now.

http://www.mp3car.com/vbulletin/soft...nd-update.html