Ideally, a scientist should keep an open-mind. Once you have decided the outcome of an experiment before you start it, then you limit the opportunities to make truly startling discoveries. Bold claims require extraordinary proof. But that doesn't make them impossible.
The original cold fusion experiments were simply bad science. Pons and Flieschmann made several mistakes in their experiment and were most likely misled by seeing false signals in among the noise - which is a mistake that I made recently and is very hard to avoid until you've made the mistake yourself. They did however, most likely discover something that is not fusion but is not clearly understood to this day. Their biggest mistake, however, was in hyping their discovery before they underwent any form of peer review or before attempting to clarify their results or even reproduce them.
The article in the original link mentions "excess energy" several times - once in bold print. But energy can be produced in numerous ways - most notably a chemical reaction. Excess energy does not prove or disprove "cold fusion" - it's a start, but remnants of the nuclear reaction should be measureable - and while their experiment does produce energy, the other byproducts of a nuclear reaction are not as easily reproduced - and what has been measured is a very small signal that is barely detectable above the noise - or, as many others think, is really noise itself.
That said, enough experiments have followed that seem to indicate something is going on in their experiment that is not clearly understood. It's almost certainly not fusion, but what it is is interesting and is worth persueing.
As far as the original question "is cold fusion a reality?" I would say, no one can be sure.
There are still plenty of questions regarding
Sonoluminescence which is possibly fusion occuring in a benchtop setup that involves essentially sugar water (or mineral oil), deuterium and big speakers. Skeptical? Sure. But worth investigating because, similar to the Pons and Fleishmann experiment, there is something going on that is intriguing and is not well understood.
There is also the recently published Nature article detailing a remarkably simple aparatus that appears to perform nuclear fusion in a tabletop setting. Layman-level details can be found in this
USA Today article. For more specific details, look in the May issue of Nature.
Neither sonoluminescence or the recent fusion apparatus claims to be "cold" fusion. But both are fusion in setups that are not remotely close to a multi-billion dollar spherical tokamak reactor.