In the past I have traditionally made some "un-predictions" rather than predictions for the new year. The reason for this is that I've found it's far easier to predict things that won't happen rather than predict what will happen.

So here are my "un-predictions" for 2015:

There's been a lot of buzz on SDxCentral and others about the soap opera that is Juniper Networks. As many of you may have followed, the former CEO, Shaygan Kheradpir, resigned last year under mysterious circumstances revolving around an unnamed customer deal.

That's not all. At the same time, Juniper's technical plan for a transition to software defined networking (SDN) has been confusing, with CTO Pradeep Sindhu appearing a little conflicted on the strategy. Juniper is also being investigated under the auspices of the Foreign Corrupt Practices Act. Oh yeah, all this is happening while the company remains under restructuring pressure from investors, including activist hedge fund Elliott Management. Good times, eh?

Bottom Line: Juniper's kind of a mess. Its SDN strategy is murky. The investors are raising their pitch forks. None of this will be fixed in 2015.

Okay, this is kind of a double negative. So I guess I'm making a prediction: Cisco Systems CEO John Chambers will finally leave. If I'm wrong again, it's not my fault, it's his.

In 2014, I wrote a (wrong) story that John Chambers would leave Cisco before the end of last year. Silly me. If you think I'm causing trouble, I'm not. Chambers actually floated the concept as far back as 2012. Where there's smoke there's fire. Everybody is buzzing about this, from Wall Street to San Jose, California. And I think eventually pressure will grow in 2015 for some major change.

Chambers has had an epic run at Cisco and should move on while it's a natural time for transition. With the heyday of the Ethernet switching market winding down and moving to the software-based cloud networking world, it's a good time for somebody else to take over. Maybe somebody with more software chops? Chambers should take the cue from Michael Jordan circa 1999 and retire at the top -- and not consider any moves to the Washington Wizards.

Everybody's moved on from hyping SDN to hyping SDN's more understandable cousin, network functions virtualization (NFV). NFV is a powerful, real trend, but it's going to happen in a gradual smooth way, not in one big wave.

NFV is going to shake up many of the communications markets and subsectors, including WAN Optimization, Applications Delivery Controllers (ADCs), and firewalls. But the fact is that many of these players have already started shifting their architectures from hardware to software and now. The next step is to help their customers make the transition. The trend will continue, but in a slower, more evolutionary way.

This is an easy one, especially after AT&T recently announced that it will reduce capex spending by as much as 20% in 2015. The service providers have taken SDN and NFV and basically used them as an excuse to reduce spending. Apparently they'd rather blow huge sums of cash on mobile spectrum.

Aggregate global service provider (SP) capex increased in the low-single-digit percentages in 2013 and 2014 but is forecast to decline by 2% CAGR from 2015-2020, according to research from MKM partners.

"This is not an industry environment that will lift all boats," wrote MKM partners managing director, Michael Genovese, in a recent research note covering secular industry trends and capex spending.

This is a great point, and one that bears careful watching in 2015. The Capital spending shift will benefit the concentrated batch of companies that are able to provide value in the new software-based architecture. Once this shift plays out, it's going to be in place for many years.

The cloud consumes everything. Think about it: your time, your data, and your money -- it's all going into the cloud. So what's to stop it?

If you think about the industries that have been slow to adopt cloud -- think of the music industry and the service-provider industry -- that's been a recipe for disaster. I find the story of "Interview" and Sony Pictures hacking disaster as a parable for the cloud. That company's entire business has become cloud dependent, whether it entails locking down its data and resources or distributing the film via digital media services.

As we move into 2015, the cloud will become more powerful across all industries. And the technologies that power the cloud will become more valuable and more powerful than ever before.