Estimated development time - Printable Version +- Kodi Community Forum (https://forum.kodi.tv) +-- Forum: Development (https://forum.kodi.tv/forumdisplay.php?fid=32) +--- Forum: Kodi Application (https://forum.kodi.tv/forumdisplay.php?fid=93) +--- Thread: Estimated development time (/showthread.php?tid=129258) |
Estimated development time - laurens - 2012-04-18 Hi everyone, For an upcoming project I was hoping to get some input from developers more experienced than me on how long the following features will approximately take to implement (in hours). While I have been diving into XBMC's source code, it has been to short of a time to accurately estimate the amount of effort it will cost and I do need to start making decisions here. - Internet browser through Awesomium - Implement categories as an alternative to plain directory browsing for videos, photo's and music - Integrate social media (let's say just Facebook for the time being) to the program guide ('like' a show for instance) - Add overlays to videos. So for instance, when watching the Robin Hood trailer, display a clickable link to NetFlix. If there are any obvious hard parts I would love to hear what they are. Thanks! RE: Estimated development time - RockerC - 2012-04-18 (2012-04-18, 09:18)laurens Wrote: - Internet browser through AwesomiumIf you are planning on coding an embedded overlay web browser for XBMC integration then read this before you dig in deeper into Awesomium http://www.ogre3d.org/tikiwiki/Awesomium That explains that the newer versions of Awesomium is unfortunately no longer open source under a GPL compatible license as it was before. So you might want to checkout Berkelium instead which is initially as fork of an older version of Awesomium from back when it was open source http://berkelium.org https://github.com/sirikata/berkelium I also suggest you checkout ideas from MeeGo TV Browser project for web browser GUI interface and interaction on a large screen television http://wiki.meego.com/TV_Browser#MeeGo_TV_Browser http://meego.gitorious.org/meego-middleware/meego-tv-browser It would also be very nice to have the Boxee HTML5 fullscreen video player and JavaScript interface controls to enable HTML5 addons in XBMC https://github.com/Boxee/html5-fullscreen-player https://github.com/Boxee/boxee-js-api https://github.com/Boxee/html5-examples https://github.com/Boxee/Boxee-Tools Then it should also be easier to port Boxee and Google TV HTML5 apps into XBMC addons If you combine this code from Berkelium, MeeGo TV Browser, and Boxee then you could have a perfect fit for XBMC web integration (2012-04-18, 09:18)laurens Wrote: - Add overlays to videos. So for instance, when watching the Robin Hood trailer, display a clickable link to NetFlix.That is possibly something that you could also achieve with the above embedded Berkelium web browser together with XBMC addon/skinning Adding Berkelium could mean that you could write XBMC addons for all Adobe Flash based web applications, even if they use Adobe DRM http://berkelium.org/faq.html Quote:Q: What can Berkelium be used for? RE: Estimated development time - RockerC - 2012-04-18 (2012-04-18, 09:18)laurens Wrote: - Integrate social media (let's say just Facebook for the time being) to the program guide ('like' a show for instance)XBMC developers have previously said that they will not add social media integration to XBMC's core themselves, and only want such features in XBMC as addons, (like for example the Trakt.tv addon) However they might change their mind if you instead would consider adding a social media GUI frontend framework with addon API to XBMC's core so that it will be easier to write social media addon for XBMC? It could then be a little similar to the PVR frontend/backend concept for XBMC, with a common social media frontend framework in XBMC that others could write addons for, with common interface for log-ins, etc. That way none of the actual social media addons for any specific social media site would not be inside XBMC core, and skinners could then just choose to skin support for such addons or not. Having a common social media GUI frontend framework and API would mean standard size of "like" buttons, star ratings feedback, etc. and windows for adding text like "this is a must see kickass movie" You could that way have different third-party social media addons in XBMC for Trakt.tv and ShareThe.TV, as well as Facebook and Twitter, etc. all only using this common API and GUI framework I guess that the same framework and API could also be used for submitting simple star ratings and rankings to thetvdb.com and themoviedb.org after viewing if the users have a user account for those sites. As well as content recommendations from Synopsi.TV, Flixster, and GetGlue based on what you are watching under your profile in XBMC, similar to Boxee but not only using social connections. Something similar to Last.fm music scrobbling and movies recommendations in Netflix http://betakit.com/2012/04/16/synopsitv-launching-to-improve-movie-and-tv-recommendations http://www.engadget.com/2012/04/18/google-tv-tv-and-movies-app-update/ Boxee would then get a little competition, though it would be nothing to stop them from also help develop this common social media framework and API for XBMC, as could probably get ideas from their GUI RE: Estimated development time - laurens - 2012-04-18 Hi RockerC, Thanks for the swift and detailed response! It has given me a lot to consider and think about. Would you be in any position where you could comfortably guesstimate the amount of effective hours that would go into developing these features? Thanks! RE: Estimated development time - RockerC - 2012-04-18 (2012-04-18, 13:56)laurens Wrote: Would you be in any position where you could comfortably guesstimate the amount of effective hours that would go into developing these features?No sorry, that is something that XBMC team developers will be able to inform you about. I have by the way updated my two posts above with some more information and ideas. RE: Estimated development time - davilla - 2012-04-18 you're kidding right 4-6 months. That's right, months not hours. RE: Estimated development time - laurens - 2012-04-18 Hi Davilla, Is that based on 8-hour workdays, at 5 days per week (if so, I was expecting that myself) or are we talking about 4 to 6 months in actual hours? That'd be a big difference. Thanks! RE: Estimated development time - davilla - 2012-04-18 Your 1st mistake is trying to generalize 4 major coding projects into a nice neat N-hour coding project. Sorry but real-life complex coding projects do not not work that way. There are several unknowns and several design decision points that are driven by platform dependencies. This is not a typical comp-sci project that you can wrap into a N-hour coding project. You factor in an overall amount of time, plus extra to try and cover the unknowns and opps. Most likely, your first attempts will work but need a second refactor to make it viable and robust. So, 4-6 months, start to finish. It's up to your coders if they feel the need to put 24/7 into it or some other ratio. RE: Estimated development time - RockerC - 2012-04-19 Would also be very nice to have Boxee HTML5 fullscreen video player and JavaScript interface controls for HTML5 addons in XBMC https://github.com/Boxee/html5-fullscreen-player https://github.com/Boxee/boxee-js-api https://github.com/Boxee/html5-examples https://github.com/Boxee/Boxee-Tools Then it should also be easier to port Boxee and Google TV HTML5 apps into XBMC addons |