Oracle shakes up database release naming convention: The Oracle database has long been and remains the market-share leader. Now the company is making a major change to its database release numbering structure, one that could have mixed results from a marketing and customer-satisfaction perspective.
In short, Oracle will move to an annual release cadence with releases named by the year, as Mike Dietrich, master product manager for database upgrades and migrations, explains in a blog post:
In my own words we basically rename the patch sets and name them what they were since years: Full releases. This means, Oracle Database 12.2.0.2 will be Oracle 18. And Oracle 12.2.0.3 will be Oracle 19. And so on.
Therefore there won’t be any Oracle 12.2.0.2 anymore – and obviously no Oracle 13.1 followed by Oracle 13.2.
Along with the annual release, Oracle will issue quarterly release updates. Historically, Oracle database customers have largely waited until the second rendition of a major release before upgrading, with the thinking being it will be more stable than the first. That approach hasn't been necessary for quite a while, but it's been hard to get that across to customers, Dietrich says:
Patch sets are full releases. Patch sets were full releases for years. In Oracle 12.1.0.2 (a so called “patch set” containing only as little as 13000 fixes on top of 12.1.0.1) we introduced complete huge and important new features such as Oracle In-Memory. Patch sets became full releases since at least Oracle 11.2.0.2.
This will hopefully end discussions. Nobody has to justify to go live on the first release. There is no first release. And there weren’t first release for many years. It were full releases.
So yes, I’m happy with this change. And it makes a lot of sense.
POV: A H/T goes to the Register for flagging Dietrich's post. Oracle's previous numbering convention was favored by developers, but this won't be the first shift toward a more marketing-minded approach, nor the first shift that takes place before the dreaded Release 13, says Constellation VP and principal analyst Doug Henschen: "Software release 13.xs are about as popular as 13th floors in office buildings."
But it's not clear that a revamped numbering convention will be enough to spark faster uptake of new database releases, Henschen adds. "Whatever the release designation of the database might be, organizations will continue to consider the lifecycle of holistic deployments, not just the latest features introduced with each new database release."
Kohl's continues evolving toward multichannel: It appears that large retailer Kohl's is having some luck weathering the competitive storm clouds generated by Amazon, reporting quarterly results on Thursday that contained positive news regarding online sales as well as improving synergies with its brick-and-mortar stores.
At Kohl's, online demand sales rose 19 percent and 31 percent of those orders were picked up in stores. (Kohl's also ended the quarter with 1,154 stores, up from 1,150 one year ago.) In addition, Kohl's smartphone app is looking pretty sticky, with 66 percent of its online traffic and 42 percent of online revenue coming through that channel. Overall, Kohl's revenue dipped very slightly to $4.14 billion, but net income shot up 49 percent to $208 million.
Kohl's goal is to continue driving foot traffic toward its stores by leveraging online ordering and in-store fulfillment, which it says leads to significant upselling opportunities. CEO Kevin Mansell said it is continuing to open new stores this year, but they are "very different" than past designs. There also shouldn't be any news regarding store closures this year.
POV: Traditional department stores like Kohl's have been some of the hardest-hit by Amazon's ability to discount, offer a massive selection and deliver a pleasurable overall customer experience. While Kohl's has plenty of work ahead of it, the company at this moment seems like a good example of a legacy business making the right, if sometimes painful steps, toward transformation and ultimately survival.
DARPA funding research into AI that explains itself: The U.S. Defense Advanced Research Projects Agency (DARPA) is bankrolling research with the end goal of creating artificial intelligence systems that can explain to humans how they arrived at a given result.
The Palo Alto Research Center, a subsidiary of Xerox, has been awarded a DARPA contract to create a system called COGLE (COmmon Ground Learning and Explanation), as the company describes in a release:
The key idea behind COGLE is to establish common ground between concepts and abstractions used by humans and the capabilities learned by a machine. These learned representations would then be exposed to the human via COGLE’s rich sense-making interface, enabling people to understand and predict the behavior of an autonomous system.
It's all about gaining crucial context for an AI system's conclusions, as well as to develop trust in the system. COGLE will first be developed in conjunction with an unmanned aircraft system but its design will be applicable to other types of autonomous systems later. Carnegie Mellon University, West Point, the University of Michigan, the University of Edinburgh and the Florida Institute for Human & Machine Learning are also involved in the research.
POV: It's not just DARPA that's interested in AI that is explainable, as financial organizations also insist on recommendations and automated decisions that are explainable for reasons of legal and regulatory requirements, notes Constellation VP and principal analyst Doug Henschen.
"When it comes to decisions on lending, risk, pricing and claims, financial organizations have to avoid black-box approaches because they have to be able to explain to regulators why certain decisions were made," he adds. "Regulators want to ensure there isn't bias or bad math behind the scenes. The challenge of ensuring human transparency in decision systems will become more challenging as organizations seek to rely on machine learning or cognitive systems that constantly adapt to the results, as expressed in data, of each new decision."