Well, something had to go wrong. The Apple (NASDAQ: AAPL) iPhone got pretty good reviews. The lines were long. AT&T's (NYSE: T) CEO gave an interview about how great the whole thing was.

But, customers who bought the phone are finding that it takes some time to activate the AT&T service that operates the phone. And, they are angry. The time lag for setting up the phone so it actually works affects both current AT&T customers and people moving their service from other companies, like Sprint (NYSE: S).

AT&T and Apple may be trading comments about which company is to blame. As one user posted: "Same here (in NYC), it's been about 10 hours for me. AT&T said (after talking to 4 agents) that it's Apple's job to activate the phone not theirs," added user PeteTech.

No product launch is perfect, and it's safe to assume that, once the initial flood of iPhone buyers has passed, getting the handsets to work will not take so long.

Maybe you can get better activation time if you offer to pay another $500 in addition to the $500 for the phone.

Douglas A. McIntyre is a partner at 24/7 Wall St.

A recent article by Melly Alazraki examines the rapid advance of voice-recognition technology and its application to search utilities. As areas of speculative investment go, this is a great one! I suggest getting on the voice-recognition train right now because it's going to take off and once it does, there will be few stops between here and the top.

Now here's what you want to look for because this is being utilized successfully in warehouses and distribution centers around the globe. You want to find the companies who are developing voice-recognition programming which is "normalized" by the user at time of inception. Simply put, the best voice-recognition applications are user specific in the recognition of voice patterns.

On the industrial side, warehouse and DC workers spend a brief time introducing themselves to the voice-recognition programs by "installing" their own voices into the programming patterns. The applications come to recognize each employee's individual voice and speech patterns. Special nuances such as a lisp, teeth clicking, slurring or a nasal tone are cataloged as specific speech characteristics upon which the programming can rely for accuracy and validation of information and the individual. The industrial world has found that this allows the systems to adapt to any amount of accents, dialects and colloquialisms. In many cases these programs will even act as interpreter between employees and allow individuals who speak entirely different languages to interact effectively.

Please do look into voice recognition as your next cutting-edge investment. Check on the industrial side of things to get the real picture about what is working. Newly developing voice-activated search is already working at our house. I used it just moments ago. I told our two year old to go get her shoes... oh, never mind, that's something altogether different.

Microsoft (NASDAQ:MSFT) is claiming "no harm no foul" in a hackers dream scenario which has come to light. It seems Windows Vista may allow would-be evildoers to whisper sweet nothings in your computer's ear and subsequently shut it down and/or delete material. An article at Red Herring exposes an apparent voice recognition loophole in the new Windows Vista operating system. The claim is that it has been shown that a hacker can invade the system... for or with a song.

Microsoft has downplayed the glitch, claiming that it's a small problem and it has similarly been encountered before. If that attitude is intended to make consumers feel better about possible system breeches, I'm afraid that Microsoft has another thing coming. Perhaps MSFT hasn't heard that PC security tops the list of consumer concerns in the IT world. Perhaps I need to send them a scathing letter (place sinister chuckle here).

We need to also consider how this Vista voice recognition glitch may be exploited further down the road. In light of the heavy usage of P2P BitTorrent type technologies for music sharing and similar, what is the potential for nasty audio hacker attacks to become widespread and destructive? If Windows Vista can be manipulated by virtual attacks placed within or alongside a song then just how far can audio-borne hacking go?