Discussion Locked

Making Enterprise Desisions - Software

0 helpful
4 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.
We as a multi-state workers compensation insurance compane are faced with the choice between two competing software products that manage the claims process for different sections of the US. On one hand is the Oracle formes 3.0 product that has extensive functionality but not multi-state capability. It is concentrating on the state of CA. It will need to undergo a two part upgrade to bring it up to the latest tool set in Oracle and use Oracle 8i as the backend. The developers will need to learn how to design windows GUI screens that are user pleasing and not overbearing. They will also need to train 250+ staff members in the use of these upgraded screens and how to use a mouse to navigate them (a new experience). On the other hand is a software product that is already using windows GUI standards and built in PowerBuilder 4.02 and upgraded to 6.5 currently. Plans to upgrade to 7.0 are pending a company decision on which system should be the enterprise system of choice. The backend dayabase is Sybase for this system. The estimated cost of converting to Oracle8i for this PB system is $500,000.00. This system operates in 13 states. How would you go about deciding which system to choose? All suggestings are appreciated. Thanks!

4 replies

by brianharris

Its been a while since I've seen Oracle development tools, but I got the impression from my experience and comments of other developers that Oracle is a terrible development environment. Learning it to the point of being productive will be fairly difficult. Now things may have changed over the years, so I don't if this is still the case. Oracle databases are great though.

I also have the impression that Power Builder isn't nearly as good as it used to be for developing Windows apps. And since its market has gone down quite a bit, I'd be worried about the maintenance.

$500K for a conversion is too much. This just shows me that the development of the system wasn't planned out very well. That would cause me worry about other aspects of the system. Doesn't anyone know how to do proper n-tier system development?

One other suggestion I might make if you have the time to do it would be to bring in a consultant to see if you could have a system custom built for less than thesechoices. You might not but if you're spending a big chunk of cash it might be good to find out. A piece of advice, stay away from the Big 5 consultants. They almost always over charge and under deliver.

Good luck!

by al macintyre

Differnt people focus on different challenges & may often be constrained by choices made at other levels of management.

One issue you have not addressed is the ability of the final package to weather the storms of laws rapidly changing, which means you need to get your data base rules updated promptly since the governments often give you very little notice to be compliant in the latest new thing you need.

Implementing this is much more practical & inexpensive if you are using a package that already has the rules for a particular state, or able to support any states.

When I started at my current employer, the dingbats who had designed the prior payroll system had a huge file with duplicated fields for 3 states that we had employeesin, then we opened a location in a 4th state & the system could not process it. The data needed to be structured so that state is part of the definition.

If you are in a highly modified reality & the base software package gets an upgrade, what does that do to your highly modified version? You have to think about that process in advance. Some software gets upgraded like every 6 months or more often from the software vendor ... ideally you want the impact on you to be no more than a weekendto weather the storm. That is just not doable from some suppliers.

by al macintyre

I suggest you check out IBM's UDB, LANSA, and ERROS. These systems let you define your business rules whatever way makes sense, then deliver GUI to users in whatever is the hot interface of today ... Java, CGI, HTML XTML whatever - 3 years from now it will be totally different. Yes you can become an expert at writing CGI in C or whatever language, but down the road it will need to be a different interface, which is why it makes sense to use something like LANSA that takes your legacy code &converts it to GUI face (Win or Browser) & as the needs evolve, it delivers in the desired form.

IBM UDB is available on more platforms than any other data base language. Here is the e-server link because that is what I am on, but from there youought to be able to get to other platforms.
http://www.as400.ibm.com/db2/db2main.htm

LANSA runs on many different platforms.
http://www.lansa.com

ERROS is purely AS/400.
http://www.erros.co.uk/

by garymore

You may want to revisit the entire process flow of ?managing the claim process? for different section of US and to identify how much difference before talking about technology solution. This business process is a service to the customer but is a cost to the company. So how much improvement you can ensure these two suggested solution could help company to generate higher profile. Or, you may want to tell us more about what is your business need? Technically there are pros and cons in both Oraclesolution and PB solution because it is designed with common business requirements. You may have different requirement in conducting your business. So the implementation is totally depending on the business strategies. I suggest you to conduct a business risk analysis between business needs and technology functionality. If I make decision, I will evaluate total functionality can provided by both solutions compare to the total spending required (including conversion, customization, upgrade, training) I had work in text base screen enterprise system and migrated to GUI base screen enterprise system. During the transformation of enterprise system, everyone in the organization is facing any kind of circumstances of change. It takes about few years to stabilize the enterprise system and building the new knowledge and skills on the new business model. If your customer base or data volume is very high than $500,000 is all right otherwise technically is too much. Try to compute how much cost to convert one-customer details. You may need to know this because switching of database platform is very complex in a high data volume and planning of implementation across different platform and remote office in different states. To do this, you need a data consultant or data architect expert. Good Luck.

This discussion is closed to new replies.