Date: Fri, 17 Nov 2000 22:11:44 +0000 From: Lambert Schomaker To: tc-11 Subject: title issue / food for thought Dear TC-11 community, below is a summarized inventory of reactions (thank you). Food for thought over the weekend! Best regards, Lambert Schomaker .................................................................. Reaction #1 Why not consider the ICDAR, IJDAR convention: "TC-11: Document Analysis and Recognition" .................................................................. Reaction #2 To keep with the very sensible definition you have: > [INPUT: pixels or planar coordinates]==>[OUTPUT: text and layout > interpretation] I would suggest "TC-11: Text Image Processing" (without the qualifier) because the mention of "Text" sets us apart from TC-10 and the mention of "Image" sets us apart from those whose input is ascii. .................................................................. (LS: Although these two proposals are reasonable, a comment on #1 and #2 is that the wording stresses the traditional OCR approach and underexposes the growing field of pen-based/hand-held applications, the areas of license plate recognition, signature verification, text recognition in streaming video and other innovations in machine-based reading. A term like 'Reading', on the other hand, is more general.) .................................................................. Reaction #3 "TC-11: Formatted text interpretation?" "TC-11: Text-image analysis?" .................................................................. Reaction #4 "TC-11: Theory and Applications in Text Processing" seems good to me. .................................................................. Reaction #5 "TC-11: Reading Systems" is my favorite. ... Reading is exactly what we intend to do. This word covers the Signal Processing approach ...(upto)... the simulation of human tasks. (...) With this name, the relationship between signal and image processing and the new technologies of communication is obvious. Converting a paper document into XML is no more an application but a natural goal for the research. .................................................................. Reaction #6 ....you raise the worry that TC-11 research areas may be confused with those of (Computational) Linguistics and/or Information Retrieval (or Data Mining). With "Text Processing" in the title, this is indeed possible. So, why not choose: "TC-11: Theory and Applications of (Machine) Reading Systems" .................................................................. Reaction #7 I would go with: "TC-11: Reading Systems" Indeed, Reading encompasses all PR trends from computer science to cognitive fields using statistical methods, structural/syntactical methods, connectionist models, stochastic approaches, reading models from perception, etc. (...) I think theory is a must to produce a solid basis to our practical problems. Applications will certainly drive theory. The former word will attract and consolidate our role as per technology transfer to industry. Following this line, I would suggest: "TC-11: Reading Systems: Theory & Applications" ..................................................................