Re: Help with database strategy

This WebDNA talk-list message is from

1998


It keeps the original formatting.
numero = 18100
interpreted = N
texte = >Hi Marty - Go Relational! > >I would suggest busting your data up by catagory and having relational >fields. For example: >Marty's suggestion is the best approach. In spite of the tendency of WebCatalog's speed and performance to dazzle, it is always best to stick to proven fundamental database design and use the tools within WebCatalog to accomplish relations. If you are unfamiliar with these, don't be put off by the code - do a simple relation first to see the pattern; it is simpler than it may first seem. Lookups and embedded searches can do a lot for you, and make maintaining your datafiles so much simpler.A fundamental advantage of using related files is that when the ugly spectre of scope creep from a client rears its mighty head, a properly designed database structure will be your very best ally.A case in point - Hewitt Consulting is designing a complex Voters Guide and Elections Reporting system for the State of WA, Secretary of State (in prototype mode). We were faced after our last client meeting with a significant scope enhancement:A County with a given ballot may have 100 records associated with it, from 2 different databases.The client wants a single form to be able to display and update all these records from 1 screen, with a single update button effecting changes.We use arrays, loops, and the REPLACE function to achieve this result, all the while maintaining the integrity of the original relational database structure. It is fast, and easy for the user, and auditable (which is a fundamental requirement of the system). Without a well-designed relational structure, we would have had to probably rebuild the entire database, and compromise the complex reporting requirements which have already been programmed.It is worth the learining curve.Best of Luck.Larry Hewitt Associated Messages, from the most recent to the oldest:

    
  1. Re: Help with database strategy (Larry Hewitt 1998)
  2. Re: Help with database strategy (Marty Schmid 1998)
  3. Re: Help with database strategy (PCS Technical Support 1998)
  4. Re: Help with database strategy (Kenneth Grome 1998)
  5. Re: Help with database strategy (Kenneth Grome 1998)
  6. Re: Help with database strategy (Matthew Steele 1998)
  7. Help with database strategy (Marty Schmid 1998)
>Hi Marty - Go Relational! > >I would suggest busting your data up by catagory and having relational >fields. For example: >Marty's suggestion is the best approach. In spite of the tendency of WebCatalog's speed and performance to dazzle, it is always best to stick to proven fundamental database design and use the tools within WebCatalog to accomplish relations. If you are unfamiliar with these, don't be put off by the code - do a simple relation first to see the pattern; it is simpler than it may first seem. Lookups and embedded searches can do a lot for you, and make maintaining your datafiles so much simpler.A fundamental advantage of using related files is that when the ugly spectre of scope creep from a client rears its mighty head, a properly designed database structure will be your very best ally.A case in point - Hewitt Consulting is designing a complex Voters Guide and Elections Reporting system for the State of WA, Secretary of State (in prototype mode). We were faced after our last client meeting with a significant scope enhancement:A County with a given ballot may have 100 records associated with it, from 2 different databases.The client wants a single form to be able to display and update all these records from 1 screen, with a single update button effecting changes.We use arrays, loops, and the REPLACE function to achieve this result, all the while maintaining the integrity of the original relational database structure. It is fast, and easy for the user, and auditable (which is a fundamental requirement of the system). Without a well-designed relational structure, we would have had to probably rebuild the entire database, and compromise the complex reporting requirements which have already been programmed.It is worth the learining curve.Best of Luck.Larry Hewitt Larry Hewitt

DOWNLOAD WEBDNA NOW!

Top Articles:

Talk List

The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...

Related Readings:

Format question (2001) Grep Search and Replace (2002) IE 6 HTTP Post (2004) WebCatalog 2.0.1 NT beta 1 released (1997) Passing Variable (was How to add product size) (1997) New Webcatalog for Mac (1997) Oracle XE (2006) Word wrapping (1997) headers (2004) Random (2000) WC2.0 Memory Requirements (1997) TCPconnect issue (2000) What am I missing (1997) creating writefile data from a nested search (1997) Flash Stores (2003) default value from Lookup (was Grant, please help me) (1997) Looking up two prices in database? (1997) Credit card number for testing (2003) Javascript and Search Results (1998) emailer (1997)