Archives of the TeradataForum

Message Posted: Mon, 07 Jan 2002 @ 08:42:00 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

A very interesting debate this one!

Imo there is rarely, if ever, an overwhelming demand from the business community for new features in a release that is sufficient to warrant an upgrade. No matter how many 'cool' new features are contained in a new release, a simple fact always remains - that existing code will rarely/never be changed to take advantage of whatever is in a new release. What looks 'cool' to developers may have no real business benefit other than quicker/easier application development. Will the business ever give us credit for this though?

A new release is always s bit of a step in the dark. Each site has it's own unique way of doing things. NCR can't possibly test what each of it's clients does with it's software. Companies such as MS have a huge beta test program in order to iron out a lot of bugs. That just isn't possible with Teradata software. The installed user base is relatively tiny and a lot of companies don't run test machines in any case.

It seems to me that you have a straight choice - either bite the bullet and go with a new release, or adopt a wait-and-see approach and hope to get feedback from other sites. The latter approach is in no way guaranteed to avoid problematic releases. What works fine at site A might not be so fine at site B. It also depends on how adventurous a site is in it's use of the technology.

Some sites will hold off an upgrade for as long as possible simply in order to maintain a stable environment. I see nothing wrong with this, other than NCR wanting to keep all sites on a narrow release spectrum.

Paul Johnson.

	<Prev	Next>		<<First	<Prev	Next>	Last>>