Requirements

Introduction to Project Background and Client

The application requested by our client, Mia Ridge from the British Library, is a web-based tool that is going to be used by researchers and historians to identify authors of ancient documents. The real world problem that the app is trying to solve is to automate and improve the analysis process that is manually conducted by researchers.

Project Goals

The main goal of Scribal Handwriting is to provide an efficient alternative to graphological expertise in regards to handwriting analysis. Our solution should provide the tools necessary for a simpler and more accurate solution to identify age, location and authors of authors of a specific manuscript.

Requirement Gathering - User Research

To better identify our users and their needs we used two main techniques to gather data: surveys, to gather quantitative data and semi‑structured interviews to elaborate on qualitative data[1]. Surveys were conducted using a sample of ten students from UCL with different backgrounds and interviews were carried out with two real archivists and researchers from the British Library. We tried to maintain a medium to low level of technical knowledge in order to better represent the final users. The results of interviews highlighted the level of technical proficiency of the final users and their core needs, such as easy to use interfaces and responsive designs. From the surveys we extracted some useful data which we combined with information gathered from interviews carried out with potential future users - which we then used in the development of our prototype.

Summary of Information Gathering from Surveys and Interviews

Personas

Storyboards

Using both our research on the users and the scenarios, we created storyboards to highlight possible confusion or issues that might arise when adding the user variable. Thanks to their linear structure, storyboard helped us figuring out points where users might struggle with our application.

User Requirements

We gathered the user requirements using the MoSCoW prioritization technique[2].

ID	DESCRIPTION	TYPE
M1	System must recognize letters and writing styles	Functional
M2	System must allow to search similar letters based on a selection of a character.	Functional
M3	System must be a web based application.	Functional
M4	System must let the user select a letter through a cropping selection tool.	Functional
M5	System must allow user to import images to his own collection	Functional
M6	System must allow user to view imported images in his collection	Functional
S1	System should be able to account for false negatives.	Functional
S2	System should have a memorable UI	Non-functional
S3	System should have consistent navigation system in each page.	Non-functional
C1	System should allow each user to have a personal training data settings.	Functional
C2	System could scan through damaged documents	Functional
C3	System could use deep neural networks to improve performance speed	Functional

User Case Diagrams

User case diagram

List of User Cases

ID	USE CASE
SH1	Creating a new Account
SH2	Login in
SH3	Reset Password
SH4	View Manuscript
SH5	Upload Manuscript
SH6	Character search

Creating a New Account

ID	SH1
Brief description	Creating a new Account to access Scribal Handwriting
Primary Actor	Researcher
Secondary Actor	Firebase Database
Pre conditions	User is in the sign in/log in page. User select register
Main flow	User enter name, email and password twice. User select register.
Post Condition	User is register and logged in.

Logging in

ID	SH2
Brief description	User login using email and password
Primary Actor	Researcher
Secondary Actor	Firebase Database
Pre conditions	User is in the sign in/login in page. User select login
Main flow	User enter email and password. User select login.
Post Condition	User is logged in.

Reset Password

ID	SH3
Brief description	User reset his own password
Primary Actor	Researcher
Secondary Actor	Firebase Database
Pre conditions	User is logged in the Account page.
Main flow	User select reset password. User enter new password twice. User select reset my password.
Post Condition	User password is updated.

View Manuscript

ID	SH4
Brief description	User browse the manuscript in his collection
Primary Actor	Researcher
Secondary Actor	Firebase Database
Pre conditions	User is logged in the home page.
Main flow	User select View Manuscript Collection. User selects a manuscript preview image.
Post Condition	User is presented with the manuscript pages.

Upload Manuscript

ID	SH5
Brief description	User uploads a new manuscript to his collection
Primary Actor	Researcher
Secondary Actor	Firebase Database
Pre conditions	User is logged in the home page.
Main flow	User selects Add Manuscript to Manuscript Collection. User inputs a title. User select Import Images. User select images to import.
Post Condition	The new manuscript is added to the manuscript collection.

Character Search

ID	SH6
Brief description	User starts a character search on his collection
Primary Actor	Researcher
Secondary Actor	Firebase Database, Back End Flask Server
Pre conditions	User is logged in the home page.
Main flow	User selects Start New Character Search. User selects import Manuscript image. User selects image to import. User crops the uploaded image. User selects manuscript to search in. User selects options. User selects Start Search.
Post Condition	User is presented with a result page.

References

[1] - Ieeexplore.ieee.org. (2017). Experiences from conducting semi-structured interviews in empirical software engineering research. [online] Available at: http://ieeexplore.ieee.org/document/1509301/

[2] - DSDM Atern Handbook (2008). (2017, June 16). Retrieved October 30, 2017. [online] Available at: https://www.agilebusiness.org/content/moscow-prioritisation-0