Text Field Rewriter

[AnimeshSinha1309]

We will be implementing the discussion forums, but there people will start using too many short forms, and bad punctuation, capitalization, and mess up the entire forum. I will become unreadable that way. How about a JavaScript based key-press and word tracker, that would check if you are using short forms, and then ask you if you want to expand. For example, if you type in ol9, it will ask you if you meant online, and hit enter to convert. That would make a great text field auto-complete.

[AnimeshSinha1309]

Of course, we would need a great database of short forms, but our friends at school can help with that. All the short forms, and all possible expansions of each short form, with it usage popularity, so that we can arrange them properly from most used to least used when suggesting.

[omkrishna]

Hello then, its awkward writing 'hello' in a comment

Om Krishna said on a conversation in Google+:

And the idea you are talking about requires a specific meaning for each word inside a database, who will create such a large database?

[AnimeshSinha1309]

OK, So if you ask about the database, lets just let it be 3 words, and let the users expand it. For now, its r, u and ol9.

[AnimeshSinha1309]

We can also use something like GitHub flavoured markdown. Like bold, and italics, and @mentions, as well as block quotes, bullets, references, and more. There is what GitHub does: https://guides.github.com/features/mastering-markdown/

[AnimeshSinha1309]

This is what the database should look like, in my view. We can add/remove fields of course. Because we don't want too extensive a list, I think its best to work with JSON.

[GauravAnand563]

Hey Great idea Animesh. But Om is right at this-"And the idea you are talking about requires a specific meaning for each word inside a database, who will create such a large database?"

[AnimeshSinha1309]

@GauravAnand563 and @omkrishna, it is not a super large database. It maybe 250 odd entries for now, and although not perfect, it will be pretty good. But how do we expand, and how do we fill up our databases to be really good, that is a tough question and I will not be able to answer it. However, understanding clustering, and other Artificial Intelligence algorithms can be really useful. View these two links for some information (actually lots) on Artificial Intelligence. MIT Artificial intelligence (advanced): http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-034-artificial-intelligence-fall-2010/ and MIT Python (Basic): https://courses.edx.org/courses/MITx/6.00.2_2x/3T2014/courseware/38b4dec3afcc47729f80916080bf1956/ [week 5 to 7].

[AnimeshSinha1309]

About detecting strings and modifying other text fields, I have implemented that in the expense monitor's add expense window, where it automatically fills in the corresponding category (if it knows) when you fill in the head. I know its a long way to go, but that's the first step.