Please read the description carefully in Deliverables section. Please note that this project is high priority for me, so I will only award it to an individual or a team that promise to work on it full time, demoing their work to me every day and maintaining constant contact with me via Yahoo Instant Messenger (I am presently residing in South Asia so my time zone will be close to yours). I am in chat in order to answer questions. If I seem not responsive, wait for a few minutes. Otherwise, you may leave me a note saying that you will come to talk to me at a particular time in your timezone, or in Rentacoder Florida timezone, and I will do my best to show up to talk at that time. Looking forward to working with you, watcher
## Deliverables
Sentence translator and comparer specs: Program has 2 big textboxes below, left txtLT and right txtRT for texts and 2 narrow textboxes above, one under the other, called above txtAT and below txtBT for selected sentences Translation mode: original text is pasted into txtLT. It is tokenized into sentences: computer identifies possible ends of sentences using a heuristic, highlights them with a distinct color, and then invites the user to either add new ones or remove existing possible ones to arrive at a correct set. Then the user can select each sentence (hence displaying it in txtAT) and write/edit its translation in txtBT. This translation is immediately incorporated into txtRT in its respective place. Hence, for each sentence of original text (as identified during tokenization stage) we maintain a translation sentence, which the user will work to write/edit. Writing translation sentences for each sentence in the text is the ultimate goal of the translation being done. In theory, user can end up first translating sentence #10 in which case the translation will be displayed in txtRT somewhat indented, indicating that some sentences before it were not translated. Optionally, [UNTRANSLATED] marker may be displayed in txtRT in place of not yet translated sentences that fit some rule, e.g. there are some translated sentences to the right of them. The system should be able to handle paragraph breaks properly, i.e. it should contain the notion of a paragraph break as one of the objects, sort of similar to the sentence object. Incorporation of existing translations: Program should also allow direct incorporation of existing translation into the document by pasting it, in an appropriate mode, into the txtRT (txtRT, as you recall, is normally used to show the translated sentences together, i.e. the translated text). In this case, the user would tokenize the translated text into sentences, and it will be expected that the translated sentences match up with the original sentence by sentence starting from the first sentence (i.e. the translated text must have equal or smaller number of sentences than the original to match up properly) It should be possible to save and load sessions of the translation work, including both the original and the (partially) translated sentences and paragraph breaks.
## Platform
Windows XP, C#