Practical, widely used and well understood versioning scheme that immediately identifies the risk and effort of an upgrade cycle?
Nah we'll just use the year instead
Big Red has changed its database release cycle, scrapping names that see decimal points and numbers added on for an indeterminate amount of time, instead plumping for annual releases numbered by the year. So what would have been Oracle Database 220.127.116.11 will now be Oracle Database 18; 18.104.22.168 will come out a year later, and …
Given that a simple increment of the fourth decimal version could have 13K fixes along with new features, they were nowhere near a versioning system "that immediately identifies the risk and effort of an upgrade cycle", so the yearly versioning is indeed an improvement over what they had before.
"The shift between 22.214.171.124 to 126.96.36.199 introduced 13,000 fixes as well as “huge and important new features such as Oracle In-Memory”, he said..."
Ah yes, the small point release that could have cost you tens of thousands of dollars.
I've just spent about three years trying to get our Fusion Middleware environments upgraded from 188.8.131.52 to 184.108.40.206 and what started as a pro-active upgrade in 2014 became one trying to bring us back into support in 2017. The prospect of having to do that every year is so terrifying that I'm shutting down now and heading straight to the pub.
Shame to see Oracle go this way but they've only themselves to blame. I started my DBA career 22 years ago on Oracle but since them I've grown to love other DB systems, PostgreSQL and SQL Server being my favourite.
If Oracle really want to make a difference then how about they follow MS's lead and put a few more toys in the box for free rather than charging for every single tiny little option. Oh and while they're at it, how about they stop installing options without asking customers and then when you're audited claim the customer installed XYZ and then taking the customer to the cleaners for a dirty great chunk of cash! The day we got severely audited by Oracle and crucified was the day I lost my love for Big Red and decided to learn MySQL, PG and SQL Server.
Not sure if that is a deliberate provocation, but I think it unfair to compare Oracle with Netware.
If Netware had achieved its objectives then we really would not be in the mess we are in today.
(I can elaborate on that, but I have to go on-site now to sort out a networking emergency).
The article has this bit wrong, it wont act "as a revamp as a way to boost sales of database licences" because when you buy a licence (and as long as you continue to pay maintenance) you can upgrade to a later version of the same edition (Standard or Enterprise mostly) without any charge.
So for examples customers dont have to shell out to Oracle to go from 7 to 8 to 11 to 18. (this is aside the other numerous costs of upgrades of course). But for once, no Oracle cost.
Biting the hand that feeds IT © 1998–2019