you can write an OS in HTML5 and be native.
-16 karma · joined July 10, 2012
2. check allow install apps from unknown sources
3. get spywares installed.
Books aren't like software where you can release often and iterate. Think about your weekend prototype project. And how much effort it takes to make it into a "successful" start up. For books, you put all the effort before printing. And that effort to turn a prototype into a profitable product is what publishers provide you. And many publishers are really good at that.
Of course with ebooks and tablet, you can start thinking of a book as software: an interactive multimedia thingy that can be updated often. You are welcome to innovate this digital medium. But we are not there yet.
the rest is usually using HTTP properly.
our flow is like:
* does content have more than X number of words? * yes: prompt if the content has to be promoted as special "tweet" like feature * is the content tagged as Y and Z , or A, or B and C? * yes: prompt for other ways to promote the content. * does the content include video? * yes: prompt if the content needs to be featured in video playlists in a section page. * is it also tagged as D? should it be promoted in another section's hot-play list? * ...
our content people start with simple form asking for title, author, tags, and content (html). the content is parsed to see if any of our media (image, video, audio) is embedded. if our games are included... etc. And depending on the title, tags, author, content new input fields are injected dynamically.
order of dynamic fields are carefully chosen so that workflow is optimal.
we initially went with contenteditable blocks (or components). that failed miserably: database became complicated. We ended up building something similar to DOM tree in database.
start with a simple form. analyze user input. dynamically add/remove more fields.
It's just an advert of Datomic.
it's okay to learn things.