1. Lecture 4 Prototyping CS 540 – Quantitative Software Engineering

2. Software Requirements Process l Requirements Elicitation l Requirements Analysis l Use Cases l Requirements Specification l Prototype/Modeling l Requirements Management

3. Serial development is costly The Longer You Wait for Feedback, the more costs are sunk.

4. Business Analyst A sometime facilitator between customers and developers.

5. Technicians Cost of Work Today Technician New Cost of Work System Administrator Cost of System Operation Cost of System Hardware Software Training StandardsReuse Processes Tools Technology $1b/yr Developer Benefits of Automation

6. Project A: Order Reading and Analysis Software Size of Prototyping Effort: 12K lines of C Code (10% of final system module) Purpose: Find a method for order reading and analysis, applicable to variable formats. Duration and Staff of Prototype: Four people for eight months.

7. Prototyping Results 1. Final requirements based on prototype results. 2. Alerted developers' to the possibility of a having a tunable system. 3. Early evaluation of functional decomposition and performance showed bottlenecks in the dispatcher. 4. Eliminated the possibility of reusing code from another project. 5. Prototype was thrown away due to decomposition and performance problems.

8. Project B:Outside Plant Data Base System Size of Effort:5% of 500K line of source code. Purpose: To evaluate database structures for an outside plant data and to experiment with approaches to handling multiple future states of equipment usage. Duration: Three people for 15 months.

9. Experience: 1.A data base structure using hyper graph theory was invented. 2.An algorithm for handling both time-driven and event- driven assignments was invented 3. The prototype became the basis for the production code. 4. UNIX flat files were used to model loop plant by way of a directed graph. 5.The prototype showed database portability from Unix to the Mainframe.

10. TRADITIONAL PROTOTYPE BASED 40% REDUCTION 20% 80% 40% 30% 45% 25% Systems Engineering Design, Develop, Test, Install Final Development, Deployment Systems Engineering & Prototype Final Development Deployment Comparison

11. Prototypes l Plan to build a prototype l Document feedback from customer for the prototype. l Manage the feedback into the software product

12. Prototyping with Reuse l Application level development Entire application systems are integrated with the prototype so that their functionality can be shared For example, if text preparation is required, a standard word processor can be used l Component level development Components are mutually independent Individual components are integrated within a standard framework to implement the system Framework can be a scripting language or an integration platform.

13. Prototyping Benefits l Misunderstandings between software users and developers are exposed l Missing services may be detected and confusing services may be identified l A working system is available early in the process l The prototype may serve as a basis for deriving a system specification l The system can support user training and system testing

14. Prototyping Approaches Prospectus Evolutionary prototyping Throw-away Prototyping Delivered system Executable Prototype + System Specification

15. Prototyping Objectives l The objective of evolutionary prototyping is to deliver a working system to end-users The development starts with those requirements which are best understood. l The objective of throw-away prototyping is to validate or derive the system requirements The prototyping process starts with those requirements which are poorly understood

16. Evolutionary Prototyping l Must be used for systems where the requirements specification cannot be developed in advance l Based on techniques which allow rapid system iterations l Verification is impossible as there is no specification l Validation means demonstrating the adequacy of the system

17. Evolutionary Prototyping Flow

18. Evolutionary Prototyping Advantages l Accelerated delivery of the system Rapid delivery and deployment are sometimes more important than functionality or long-term software maintainability l User engagement with the system Not only is the system more likely to meet user requirements, they are more likely to commit to the use of the system

19. Evolutionary Prototyping Technologies l Specification, design and implementation are inter-twined l The system is developed as a series of increments that are delivered to the customer l Techniques for rapid system development are used such as CASE tools and 4GLs l User interfaces are usually developed using a GUI development toolkit

20. Evolutionary Prototyping Challenges l Management problems Existing management processes assume a waterfall model of development Specialist skills are required which may not be available in all development teams l Maintenance problems Continual change tends to corrupt system structure so long-term maintenance is expensive l Contractual problems

21. Prototypes as Specifications l Some parts of the requirements may be impossible to prototype E.g., safety-critical functions l An implementation has no legal standing as a contract l Non-functional requirements cannot be adequately tested in a system prototype

22. Incremental Development l System is developed and delivered in increments after establishing an overall architecture l Requirements and specifications for each increment may be developed l Users may experiment with delivered increments while others are being developed These serve as a form of prototype system l Intended to combine some of the advantages of prototyping More manageable process Better system structure

23. Incremental Development Flow

24. Throw-away Prototypes l Used to reduce requirements risk l The prototype is developed from an initial specification, delivered for experiment then discarded l The throw-away prototype must NOT be considered as a final system Some system characteristics may have been left out There is no specification for long-term maintenance The system will be poorly structured and difficult to maintain

25. Throw-away Prototyping

26. Rapid Prototyping Techniques l Various techniques may be used for rapid development Dynamic high-level language development Database programming Component and application assembly l These techniques are often used together l Visual programming is an inherent part of most prototype development systems

27. Dynamic High-level Languages l Languages which include powerful data management facilities l Need a large run-time support system. Not normally used for large system development l Some languages offer excellent UI development facilities l Some languages have an integrated support environment whose facilities may be used in the prototype

28. Choice of Prototyping Language l What is the application domain of the problem? l What user interaction is required? l What support environment comes with the language? l Different parts of the system may be programmed in different languages l Example languages Java, Smalltalk, Lisp, Prolog, Perl, Tcl/TK

29. Database Programming Languages l Domain specific languages for business systems based around a database management system l Normally include a database query language, a screen generator, a report generator and a spreadsheet l May be integrated with a CASE toolset l The language + environment is sometimes known as a “4GL” l Cost-effective for small to medium sized business systems

30. Prototyping with Reuse l Application level development Entire application systems are integrated with the prototype so that their functionality can be shared For example, if text preparation is required, a standard word processor can be used l Component level development Components are mutually independent Individual components are integrated within a standard framework to implement the system Framework can be a scripting language or an integration platform..

31. Key points l A prototype can be used to give end-users a concrete impression of the system’s capabilities l Prototyping is becoming increasingly used where rapid development is essential l Throw-away prototyping is used to understand the system requirements l In evolutionary prototyping, the system is developed by evolving an initial version to the final version

32. Visual programming l Scripting languages such as Visual Basic support visual programming the prototype is developed by creating a user interface from standard items and associating components with these items l A large library of components exists to support this type of development l These may be tailored to suit the specific application requirements

33. Visual programming with reuse

34. Problems with visual development l Difficult to coordinate team-based development l No explicit system architecture l Complex dependencies between parts of the program can cause maintainability problems

35. User interface prototyping l It is impossible to pre-specify the look and feel of a user interface in an effective way l UI development consumes an increasing part of overall system development costs l User interface generators may be used to ‘draw’ the interface and simulate its functionality with components associated with interface entities l Web interfaces may be prototyped using a web site editor

36. Mindset l Prototyping is critical system parts that cannot be effectively pre-specified l Rapid prototyping focus on partial functionality l Prototyping techniques include the use of tools, very high-level languages and database programming l Users must be involved in prototype evaluation

37. Growing from Prototype to Model

38. Case History Harbor Radar Senior Design Project

39. Project Overview l To primarily serve boaters in the Hudson River area with a 40 mile radius and inexpensive TV to display weather info, buoy data, and plotted locations l Interactive website Accessible to everyone Secure section where admin can control information displayed with ability to plot coordinates and alter broadcast information l Difficult time writing requirements and setting scope

40. Requirements l Website Design -Easy to use, clean, intuitive and clearly display maps l Reliability High reliability requested by customer Availability »Ultimate goal of 24/7, however probably not doable. 20/7 has been suggested

41. Requirements (cont.) l Performance Website will support 50 simultaneous users »Reach for goal of 100 Television broadcast(s) place little load on the system Response Time »Average of.5s l Constraints Television resolution will be an issue »Aim to be readable on a 9” screen Support both IE and Netscape, versions TBD

42. Map Module Screenshots

43. Larger View Grid Icons

44. Wrapper Module for TV Screens

45. Case Study Space Data System Source: A. Hooke, NASA/JPL A Ground Tracking Network One or More Spacecraft An Instrument Control Center A Spacecraft Control Center A Science Facility A Space Tracking Network Commodity Space Communication s Systems Commodity Space Navigation Systems One or More Instruments

46. CS 540A Fall 0646 Host - Host/Terminal - Host Dependent - Centralized Database S - Client/Server - Host Cooperative - Shared Database Case Study: Networked Computing

47. CS 540A Fall 0647 Host S - Local processing continues - Database updates stored - In business Disaster Avoidance CLOSED - Processing stops - Out-of-business Networked Computing in Action