IBM REDISCOVERY Q&A'S _____________________ Q1. What is ReDiscovery? A. ReDiscovery is an OS/2-based productivity enhancement tool that enables you to organize and manage information about files on your com- puter systems. ReDiscovery scans source information and stores pertinent data from the source and from the library management system into a portable file, called a catalog. You can search this catalog to find business func- tions, routine names, data elements--really any character string found in the files cataloged. In addition, you can correlate test cases, documen- tation, sample code, ownership, descriptions, and other such information with these files. This enhances your ability to find things quickly and easily. ReDiscovery can operate against several sources at the same time, including MVS, LAN, and TCP/IP environments. You may search one or more catalogs at a time; ReDiscovery will retrieve either descriptions about what it found, or the actual files. ReDiscovery also allows you to detect changes to files to which you've subscribed. ReDiscovery's primary value is in reducing the amount of time it takes to search, locate, and retrieve files and information necessary to perform maintenance and application development. Q2. Who uses ReDiscovery? A. These major user groups benefit by using ReDiscovery: o Maintenance Programmers can - quickly locate functions within unfamiliar code - identify all related code, test cases, documentation, etc. to a change request - answer questions such as: -- What modules call this routine and what modules does this routine call? -- Are similar changes being made by someone else in unit test or system test? -- Where can I find the source code that defines the class Employee so that I can define the subclass Temporary? o Analysts can - quickly identify components that need to be used, modified, or tracked across a large project, such as language migration, data migration, client/server development, etc. - answer questions such as: -- How many, and which are, the modules calling 'CBLTDLI' so that I can change the name to 'CEETDLI'? (IMS LE/370 change) -- Can I find all instances of date-related variables and date- formatted strings in order to plan our Year 2000 conversion? -- When the programmer changed the code, did they update the documentation as well? -- Where is 'NUM_FIELD' used in the application? o Managers can - support cross-team cooperation through the sharing of software assets and components across a single or multiple groups - surface data for generating metrics about applications - select and enforce programming and documentation standards across programming team(s) - develop and promote Reuse across applications Q3. What kind of files does ReDiscovery catalog? A. ReDiscovery accepts files that can contain requirements, specifica- tions, design, code, test cases, documentation--really anything (and everything). Q4. Does ReDiscovery just create an index of character strings, or does it catalog additional information about the files? A. ReDiscovery also catalogs meta data associated with the files as extracted from the file or library system in which they reside. These system-defined attributes can be information like: name, location, date created, date last modified, size, etc.. Q5. Is any other information cataloged? A. ReDiscovery has a user-customizable classification scheme that lets you add your own classifiers, also called user-defined attributes. For instance, if you'd like to know who originally created a file, who's responsible for it now, where it's used, or provide a brief description of its purpose, just add these attributes to a classification scheme and then add this descriptive information, or attribute values, to these user-defined attributes associated with the files. This is especially useful when using ReDiscovery to support Reuse activities in your organ- ization. Q6. From where does ReDiscovery retrieve the files that it catalogs? A. From almost anywhere. Just point ReDiscovery wherever the files reside: OS/2 HPFS or FAT file systems on your own workstation or on OS/2 LAN servers, AS/400 servers on your OS/2 LAN network, UNIX or AIX servers--even MVS partitioned or sequential datasets! It also retrieves files in PVCS or PANVALET library systems. A single catalog can contain information about files from a combination of these sources, and you can create as many catalogs as you'd like. Q7 Sounds like you need additional software to do this. A. Yes, but only those communications programs that you don't already have--in addition to OS/2. For example, if you wish to operate with TCP/IP, then you'll need the appropriate level of TCP/IP. Refer to the ReDiscovery User's Guide or Announcement Letter for complete details. Q8. What do you do with the results of a search of a catalog? A. View the files, copy them to your local workstation for modification, or print them. You can even launch your tools of choice against the files returned by a search. Moreover, you can view, copy, or print the system- and user-defined descriptive information associated with the files. Q9. How many catalogs should you have? A. It is recommended that you build a catalog against each source location. For example, when cataloging a MVS COBOL application, you may choose to build a catalog for your production source code, another for your JCL, and yet another for your documentation. Since ReDiscovery allows you to search multiple catalogs simultaneously, you can retrieve information about any part of your application. There may also be times when you're only interested in a particular portion of your application. In that case, you may choose to interrogate only the one catalog in ques- tion. Q10. How do you share catalogs with others? A. In several ways. You can place a catalog on a shared file system. If on a read-write disk, anyone can add or modify user-defined attribute values. More often, though, you'd place a catalog on a read-only disk. You could then make a copy of the catalog on your own workstation. If you need to add or modify attribute values, you can make your modifications and then automatically submit them to the catalog owner. The catalog owner reviews all such submissions from the catalog users and either accepts them for inclusion in the catalog or rejects them. You can even ship catalogs across the country or across the world for remote users to interrogate and/or access your files. Q11. What else can ReDiscovery do? A. ReDiscovery lets you subscribe to files in which you're interested and it'll notify you whenever those files are modified. Additionally, you can use ReDiscovery to help enforce standards across your organization. Q12. What is the strategy behind ReDiscovery? A. ReDiscovery is intended to enhance your Redevelopment efforts as well as operate with your 3GL and object-oriented environments. ReDiscovery reduces the time it takes to perform your maintenance and application development tasks, even as it reduces the risk of missing vital informa- tion when performing these tasks. ReDiscovery can also organize, manage, and support a Reuse environment. Q13. How do you order ReDiscovery? A. If you want to access files on your workstation, or on LAN or WAN servers, you'll need to order ReDiscovery/2. It's available in incre- ments of 1, 5, and 10 user licenses. If you want to access files on your MVS system, then you'll need to order ReDiscovery/MVS. This version includes everything included with ReDiscovery/2, as well as the facilities required to access your MVS files. The base package includes 10 user licenses; additional licenses may be purchased for a nominal charge. To order ReDiscovery, just phone 800-860-2047 Ext RD1, or contact your local IBM marketing representative. MORE Q&A'S ... Q14. Can ReDiscovery help with the Year 2000 problem? A. You can use ReDiscovery to search for and identify instances of date- related variables or date-formatted strings. You can then tag these files in the catalog to identify them for modification. You can include additional information such as: who'll modify the files, how long the change process should take, how much it should cost, when the changes are complete, which test cases should be executed--this is just some of the information that you might want to track during the course of your con- version project. Once populated, the catalog can be shared across your enterprise. Developers will understand what code needs to be changed, testers will understand what test cases need to be executed, and project managers can monitor the progress of their Year 2000 conversion. Q15. How can ReDiscovery help manage today's object-oriented programming environments? A. Typical object-oriented programming environments consist of a library of reusable parts and a visual construction tool to help you combine the parts to create an application. They also supply a browser to help you find the parts that you need. However, most parts libraries are not com- prehensive enough to satisfy all of your programming needs. That's when you either subclass the classes supplied with the tool, or develop entirely new classes to specifically address your requirements. When you go beyond the bounds of the parts libraries provided with these tools is where you can experience the same pitfalls so often encountered with traditional programming techniques--you can lose control of your inventory of parts. Since in most cases you can't use the tool's browser capability for the parts that you developed yourself, you have to be able to find and retrieve parts stored in multiple locations, often on dif- ferent platforms. This is where ReDiscovery can help. Most object-oriented programming environments enable you to copy to file, or fileout, their classes; others just store their classes in files to begin with. Because of this capability, ReDiscovery can be used to catalog both the classes provided with the tool and the classes that you develop yourself. You can finally get your arms around your inventory of reusable parts, regardless of location. After all, what good is having comprehensive libraries of reusable parts if you can't find what you need, quickly and easily, when you need it? Q16. When you modify a file that's cataloged, how is this modification reflected in the catalog? A. You periodically update your catalogs in order to keep them as current as your needs dictate. This could be daily, weekly, or even monthly depending on the extent to which files are modified or attribute values are added or changed. An update differs from the initial creation of the catalog; an update only processes those files whose cataloged date/timestamps differ from the present date/timestamps recorded in the library or file management system in which they reside. Utilities exist that can trigger scheduled catalog updates. Q17. What are the differences between ReDiscovery and SearchManager? A. ReDiscovery and SearchManager target different audiences with very different file management needs. SearchManager targets document proc- essing professionals, whereas ReDiscovery was designed to specifically address the file management needs of software developers spanning tradi- tional (COBOL) to state-of-the-art (Object-Oriented) technologies in both mainframe and networked environments. SearchManager searches textual documents for words and phrases. Its users need to "find a needle in a haystack" of documents in a variety of formats like WordPerfect and Word for Windows, among others. It also has linguistic capability; that is, searching for mouse also returns docu- ments that contain mice. ReDiscovery catalogs and indexes files. ReDiscovery users don't just want to search files for keywords, but also want to capture information about their files for their own future reference, and even more importantly--to share this information with their colleagues. Their col- leagues could be down the hall or across the world; ReDiscovery catalogs are portable. ReDiscovery addresses the specific needs of software developers across the entire software development process. The files that ReDiscovery cat- alogs can contain anything: specifications, architecture, design, code, test cases, documentation--really anything. The files don't even have to be textual in nature. ReDiscovery helps you to "get your arms around" your inventory of software assets, regardless of where they reside. ReDiscovery performs keyword and file meta data searches such as: Find all files that contain the keywords 'Employee_Salary' and 'Vacation_Earned' but not 'Holidays_Taken', in PANVALET library 'Pro- duction', that have been modified by J. Miller since 5/31/93. ReDiscovery targets software developers by allowing them to associate their own unique information with the files that are cataloged. This enables searches such as: Find all files that contain the keyword 'sort', written in 'C', that operate in UNIX, created by Department W90, that are certified as reusable. Once ReDiscovery returns the list of files that satisfy your search criteria, you can view the files, copy them to your local workstation for modification, or print them. More- over, you can view, copy, or print the system- and user-defined informa- tion associated with the files. ReDiscovery allows you to subscribe to files in which you're interested. For instance, if someone else modifies one of your files or the user- defined information that you've associated with them, you'll be notified of this activity. ReDiscovery also supports the detection and enforcement of standards in the development process. For instance, when a programmer modifies code, it may be required that associated documentation files be updated as well. ReDiscovery allows you to check whether or not the corresponding documentation files were updated.