Let's get to the dirt right away. I bought two disk arrays for work. They are not cheap. They are not light. They were thrown into boxes by the Amazonauts with about six inches of air for padding. I got two cartons like this:

They were kind enough to throw a small layer of paper on top. One wonders why they protected a half-inch of air, but didn't seem to care about the big six-inch gap on the side:

We here at ZDNet have discussed the migration from retail to online somewhat ad nauseum. Here are two recent commentaries by Jason Perlow and yours truly:

• Have PC retailers lost the will to live?
• E-commerce will make the shopping mall a retail wasteland

I live in something of a wasteland myself — central Florida — and even if I wanted to buy some gear locally, I couldn't. So, I rely on Amazon and other online retailers for just about anything important.

There is no excuse for this. Amazon, in particular, should have the quality control structure in place to know to throw some frickin' padding in the frickin' box, if you know what I mean.

I've discussed this before. Amazon isn't alone. Another online retailer which I frequent, Newegg, pulled the same thing, but not with anything quite as fragile as hard drives or disk arrays:

• Why you might not want to buy a hard drive from Amazon
• Rant: online retailers and their crappy packaging jobs

Of course, the only real difference in shipping is that we don't see how badly vendors pack when sending goods to retailers, but at least that prevents packing-related rants.

Look, we need to insist on quality from our vendors. There is absolutely no reason packing has to be this shoddy.

Frankly, I expect better from Amazon. I'm disappointed (again).

Y'all (I spent an hour listening to Cook's Alabama drawl) know I've occasionally been a teensy bit hard on Apple over the years, but that's because it's part of my job as a technology columnist to hold the company to a high standard.

Now, I'll admit I prefer Windows over the Mac's OS X, but that doesn't mean I haven't bought Apple products. In fact, I have five Macs now (okay, so three of them run only Windows). My wife and I have iPhones and iPads, and we've got a bunch of other Apple products including a complete set of 4th generation iPod shuffles in every color.

Granted, the iPod shuffle collection wasn't my idea, but that doesn't mean I'm not an Apple user. Heck, I used to head up a team at Apple, way back in the day.

Given all my grousing, I'm often asked if there are any Apple products I actually like. The answer is yes. On the following pages are five Apple products I like quite a bit. I'll tell you my favorite product at the end of this piece.

For now, let's start with one of my most used.

I never thought I'd like so-called "Chiclet" keyboards, until I started using Apple's. My wife and I needed a small form-factor keyboard to use with our media center PC (yes, PC) and we found that the Apple Wireless Keyboard was reliable, convenient, and surprisingly easy to type on.

We now have four of these in Camp David, including one that is actually connected to a Mac, used in the studio. I'm starting to replace this keyboard with the Logitech K810, primarily because the K810 is both illuminated and supports three different Bluetooth devices simultaneously, but I don't like that I have to charge it, rather than just replace the batteries. Because you can simply swap batteries, the Apple keyboard is a win.

Even though I've added the Logitech to the mix, I'm actively using my Apple Wireless Keyboards every day, and probably will continue to do so until they die.

Product page: Apple Wireless Keyboard

Let's count my Apple TV's, shall we? There are two in the gym (an original first-generation unit and a second-gen hockey puck). There's a third Apple TV attached to the entertainment center.

I hacked the first-gen unit and installed XBMC on it. We use it to watch video and play tunes from our media tank. I use it when on the treadmill, but I tend to like quiet when lifting. The second-gen unit in the gym is tuned to my wife's Netflix account, and she watches that when exercising.

I've been persuaded not to hack the current-gen Apple TV in the media center rack, especially since I have a nice XBMC-running Windows 8 machine already in the rack.

I don't watch much iTunes video on the media center Apple TV, but my wife and I use it constantly to watch Netflix. So far, the very best Netflix interface we've seen is on the Apple TV. For ninety-nine bucks and a Netflix subscription, it's a heck of a lot of entertainment that's hard to beat.

Product page: Apple TV

While Apple started shipping the seventh-gen iPod nano back around November, it's the sixth generation unit that is truly a win. With all the discussion recently about wearable computing, it seems Apple has had a sweet little wearable computing device since 2010 — and has since discontinued it.

There were two big wins for the sixth generation nano. First, it was small and had a nice little color display, so you could see where you were and what you were playing. It was the same size as the old shuttles, but with a screen. But the truly big win of the nano was it had a clip. You could easily clip it to your clothes without needing to allocate a pocket.

My wife has one of these (and got me one for my birthday last year), and she wears it constantly. She hooks into Audible and has listened to a ton of books, all while getting things done around the house and office.

It's not a mistake that Apple innovated on the nano and came out with a new model. It's just a mistake that they removed the clip, while, bizarrely, keeping the far more dysfunctional shuttles in the lineup.

Product page: new generation nano (not the one I like)

Photo by: Bill Detwiler / TechRepublic

I've talked a lot over the years about how to decide between full-sized tablets and reasons not to buy a new iPad.

Even though we have a first generation and third generation iPad, I don't really like the product. I don't use it all that often, and when I do, it's usually as a helper in the studio. I don't get nearly the productivity benefits out of it that some of my ZDNet colleagues do.

Even so, when the iPad mini came out, I bought one. I figured that if I was going to write about the darned thing, I ought to use it.

And, guess what? I like it. No one was more surprised than I was. People were complaining left and right about the iPad mini's lack of a Retina display, but it's never bothered me. Now, let me be clear here: my wife and I both have Nexus 7s (which we love) and Kindle Fires (she uses it, I don't). So adding the iPad to the mix seemed, frankly, overkill.

But the fact is, it works and it works well. It's all about the size. It's easy to use as a scrap of paper (I tied Penultimate to Evernote), it makes a great bed-reader, it's handy to bring into the studio, it doesn't take up too much space, it's light, and it doesn't require a separate briefcase to pack when going out. The full-sized iPad, by contrast, is like a small ultrabook once you add a cover or a keyboard.

The iPad mini is just exactly what a seven-inch tablet is meant to be: a replacement for the small, convenient notepads we've all had and loved. The nice thing is that all my iOS software also runs on the iPad mini, so it's pretty much grab and go.

There's an iOS device I actually quite like. I'm as shocked as you.

Product page: iPad mini

Sit down. This is gonna take some 'splaining. Here's the thing: my favorite Windows desktop computer is the Mac mini. I explained it briefly in my DIY-IT gift guide last December and back then, we owned two of them.

We've since bought two more, and I've installed Windows 8 on three of them. As I said in the gift guide, I've been hard pressed to find another box this small, this powerful, and this inexpensive. It takes far less desk space than even a laptop, it has a ton of ports, with SSDs it can be fast as all heck, and it's relatively inexpensive for a 16GB machine.

I buy the server version, which comes with two drives. The biggest issue is I have to set up Boot Camp and then install Windows 8, but that's not a whole lot of work.

With the exception of the studio machine, which runs Mac OS X Mountain Lion, I never boot the Mac minis into the Mac side. They've been rock-solid Windows machines and I don't have a single complaint about them.

In fact, I'm thinking of pulling the trigger on a fifth. I'm running an old Zotac in my private office, and I'd like something quite a bit faster and more capable.

Of all Apple's product, the Mac mini running Windows 8 is, by far, my favorite. Now, would I buy these if I had to run OS X on all of them instead of Windows 8? Probably not. I find OS X a tedious chore to use.

Product page: Mac mini

The company provided me with a master unit and six cameras. I paid for the electrician out of pocket. Before I found the Alert system, I had originally wanted to set up nine cameras, but the Logitech system maxes at six, in any combination of indoor and outdoor cameras.

Although I was slightly disappointed that I couldn't string cameras everywhere without limits, the six cameras do cover the entire perimeter field of the building. Each camera has a 135-degree field of view, and while you can digitally zoom, I left them all at full, wide-angle and we are able to see everywhere on the property.

Let's start with the basics

The video-over-power technology works, and works quite well. Quality video over internal power is surprisingly solid, regardless of whether or not appliances like dishwashers or clothes dryers are running.

The system booted up and routed video properly from the first. It runs just fine. Since that was the biggest question mark, it was nice to see the technology perform with rock-solid reliability.

That's not to say there aren't some issues. There are, but nothing that can't be mended in subsequent software updates. The core of the system just simply works — and that's a darned rare thing to say about any networking product, let alone one that mixes video, power distribution, and the Internet.

As I mentioned in the previous articles, I only installed outside cameras. Each camera has its own array of infrared LEDs that are designed for night vision.

To my considerable surprise, this works amazingly well. I have a relatively large yard (land is pretty inexpensive in Florida), and the cameras illuminate all the way to the street in front and to the fence line in the back. Everything is in black and white at night, of course, but we can see the entire yard in pitch black, perfectly.

Daytime video quality is also quite good. It's not full HD, but 960x720 isn't something to sneeze at, especially when you're pumping six of those feeds through the building power. I haven't noticed any lost frames.

It looks more like 15 frames a second than 30, but even so, you can easily see someone walking up or driving by without any loss of action or fidelity (at least during the day). At night, you can't make out facial features, but you can see if there's an animal or a person walking around.

Recording

The system is set up to trigger recording on motion. You can select motion zones, so recording only happens if there is activity in certain zones. One camera in our back picks up street traffic from about two blocks away through the very edge of its view interface, so I turned off motion sensing for that small zone.

While we're talking about the motion sensing system, I should mention that the Alert Commander does offer email alerts and pop-up alerts. Since there's almost always some motion outside, I turned these off, but I can see how they'd be helpful, especially with indoor monitoring.

The system first records to 2GB microSD cards installed in each camera. If you want a bigger card, you can load up to a 32 GB micro SDHC card in each camera.

The Logitech Alert Commander software downloads the video from the cameras' microSD cards to a location on your computer, so there's another copy of the video available. I added a third backup. I've got a script that monitors the backup location and uploads the video to a remote cloud backup server. Each motion video clip is a simple MP4, so you can manipulate and examine the recorded video using standard tools.

The Commander software allows you to set a maximum amount of storage, and then deletes older recordings. At about 2GB a week, a relatively small amount of storage can store pretty nearly a year without blinking.

Installation

Installation is, in theory, quite simple. Certainly for an indoor camera, all you do is plug a cable into the camera and into the power brick, and plug that into the wall.

Logitech recommends a similar procedure for outside cameras. In fact, on the instructions, Logitech has a somewhat silly illustration showing a cable running from the outside camera, down the side of a building, to an exposed power socket. While this might be easy installation, having the camera's power right under the camera kind of defeats the whole security concept.

That's why I had to involve an electrician. First, my house has unusual wiring. As I've mentioned before, when we bought this house, it was a fixer-upper in the worst way. We pretty much gutted it and rebuilt it to my geeky specifications. As a result, we've got a power infrastructure normally more suited to a small data center than a house. Since I operate my office from home, that was a necessity.

Second, though, I didn't want to expose all the camera wiring both to the outside elements and to potential threats. The electrician mounted all the cameras to the overhanging soffits around the house, then ran the cable from the camera into the attic to the interface brick. Those power-interface bricks were installed in an array in the attic, and then ran on their own dedicated circuit back to the circuit breaker panel.

The HomePlug system is a security win as well, since the wire that extends outside the house to the camera isn't on the internal LAN, it's effectively firewalled into the camera's own private HomePlug network.

Since I needed everything installed to code, I used the same licensed electrician we used to run our power infrastructure and GigE to install the Alert system. You might not need an electrician, but keep in mind that if you're touching the circuit breaker box, some ordinances require licensing to do the work (and, of course, if you don't know what you're doing, you could be in for a nasty shock).

While we're on the topic of installation, I need to point out that you can't plug the cameras or the central hub into a UPS. Here in Florida, we have many power fluctuations, and so everything (including our crock pot) has a UPS between it and the the power grid. Logitech claims the cameras have built-in surge suppression, but we had one power failure recently which corrupted one of the SD cards:

I was able to reformat the card remotely and continue operating, but I do wonder how these cameras will perform over the long haul in the face of central Florida's weather and its impact on the power grid.

Next: Odd remote behavior and final recommendations.

Odd remote behavior

A big part of having an Internet-centric, mobile-enabled video surveillance system is Internet and mobile. It's here that the Logitech Alert system has some odd weaknesses.

First, the good: you can access your cameras (and, with a yearly fee, your recordings) from anywhere in the world via either a Web interface, an Android app, or an iOS app.

Both the Web interface and the Android app display what's happening in near real-time. They're each delayed by about a half second from the actual real-time display of the internal Commander software and actual meatspace activity. That means that if you're watching a person walking up to the house, the subject is probably about one step closer to the house in reality than what you're seeing.

For access all over the world, that's not a bad delay at all.

And then we get to the iOS version. I tested this on a third generation iPad and an iPad mini. With both of these, what's shown on the iPad is anywhere from 20 seconds to about 90 seconds behind real life. If a car drives by, you can say "one-Mississippi, two-Mississippi" a good 20 to 30 times before the car drives by on the iPad.

This doesn't render the product unusable on the iPad, but it's certainly not ideal. And given that both the Web interface and the Android version on my Nexus 7 displayed in near real-time, it's not my network connection or even the delay to Logitech's data center and back. It's the iOS app.

I have one more possible deal-breaker complaint about the remote implementation, and this applies to Web, Android, and iOS equally.

Monitoring remotely is time-limited. What I mean by this is you start monitoring, and after 5-10 minutes, the monitoring stops. You're actually presented with a display arrow that invites you to tap it to start monitoring again.

This completely defeats the idea of having a browser window at work open all the time to see what's going on at home, or a Nexus 7 sitting on the night table to monitor the baby. You can check in and see what's going on, but you can't just keep an eye on things.

Presumably, Logitech thinks you'll get an email alert, and then want to see what's happening. A more cynical view is they're limiting the time the connection is live to save overhead at their data center.

In any case, it's a serious flaw and Logitech needs to step up and fix it before this solution can be a truly credible remote monitoring solution.

Update: the Web interface has stopped showing one of the cameras claiming "Error getting camera settings", while the Commander PC-based interface works fine. I haven't had time to reset everything to see if it comes back online, but remote monitoring is definitely a bit more troublesome than local monitoring, which is rock-solid.

Wish list

I've been using the system for a few weeks now, and with the exception of the remote monitoring functions mentioned above, I'm quite impressed. I did find a few features I would have liked to have seen, and so here's a short wish list for future versions.

Many security systems have a way for the system to toggle between each camera in rotation in full screen. While Logitech Alert can display all six feeds simultaneously in a grid (which, by the way, is slick as heck), I'd like to see (or haven't found) the option to simply toggle between all six views in order, continuously.

The other thing I'd like is an instant 15-second or 30-second rewind. I'm probably spoiled by my Tivo, where you can easily jump back 15 seconds to replay what you just missed, but I found I really wanted that feature on the Logitech system. I'd often see something out of the corner of my eye, and rather than going into the recording system and trying to pull up the right recording, I just wished I could hit a quick replay button and see it play back.

The rest of the story

I told you about how Jerry the electrician didn't believe this thing would work. After completing the installation, his first comment was "Wow, it works." After all six cameras came online, he was enthusiastic enough about the product that it looks like he'll be recommending it to his clients.

Maybe next time I come up with some scheme I want his help with, he won't be quite as dubious. Nah!

Final recommendations

I've always generally liked Logitech products and for my purposes — seeing what's going on throughout the property and recording what I don't see — the Logitech Alert system performs quite well.

Technically, in terms of setup, running video over power, and even night vision, the system performs outstandingly well, substantially beyond both my expectations and what my electrician thought was possible.

However, while remote monitoring works, it has a few flaws. The delayed playback on iOS is annoying and the system does have a potential fatal flaw in that you can't perform continuous monitoring remotely, which means you can't set up tablet or browser at a security desk and monitor a whole building, watch the baby from the couch, or keep an eye on kitty from the office.

Because the monitoring stream cuts off after a few minutes, I simply can't recommend this system if you want continuous remote monitoring.

With everything else working so well, I do hope Logitech remedies this one flaw in a future update. At that point, I'd be able to recommend it unreservedly.

Busted videos

Logitech has collected quite a few surveillance videos showcasing some interesting behavior. They call these Busted Videos and some are a hoot to watch. My favorites are the large black bear and the small white cat:

Yesterday, I discussed the basic premise behind building an Internet-centric, mobile-enabled, full-perimeter video surveillance system. I mentioned that I'd looked at a product from LOFTEK and discarded it, and then found a product from Logitech that had some promise.

In this article, I'll discuss the Logitech Alert system and then, in a follow-on article, I'll formally review the product. Logitech was kind enough to provide me with a full system, so I went out and hired an electrician to install it.

In fact, that's a great place to begin our story.

"No way that's gonna work." That's what my electrician Jerry said when I'd described the Logitech Alert system. Jerry's been a really big asset here at Camp David, installing all our networking cables in the attic, and completely rewiring what was a fire hazard fixer-upper into what's now part small data center and part family abode.

He looked at me with that particular "you must be kidding" look of dubiousness that skilled tradespeople often reserve for those of us who are engineers, and yet can't be trusted not to blow ourselves up.

I could see him looking more and more doubtful the more I described the product. Where he lost all faith in this scheme of mine was when I told him the video signal would travel over the power lines. Actually, that's not strictly true. He looked at me with that particular "you must be kidding" look of dubiousness that skilled tradespeople often reserve for those of us who are engineers, and yet can't be trusted not to blow ourselves up.

What got him really shaking his head was when I told him that not one, but six video signals would have to travel over the house's power, across phases, and from the farthest reaches of the perimeter to a central network node, which was then connected to my router.

When I told him I was intending to pump color, 720p HD video across the power lines, he shook his head one last time and gave me a homework assignment: "Go hook it all up inside and let me know if it works. If it works, then we'll talk."

Let's talk about the system.

The Logitech Alert system can be set up either inside or outside. If it's set up inside, you can deploy up to six cameras, what are essentially glorified webcams. I didn't do the inside thing because I have no desire to monitor the inner workings of my inner sanctum. The system can also be set up with a mix of inside and outside cameras.

I chose all outside cameras. Unlike the LOFTEK, they're fixed view only. You point them. That's what you see. On the plus side, they have a pretty wide view (135 degrees), so they do see quite a bit (although it can look pretty fisheye-ee if you set the widest angle possible).

Each camera comes with a long, thin Ethernet-like cable and what looks like a large power brick. This is actually the video-over-powerline interface that is at the core of the system's communication. Each camera also comes with an SD card, so if communication is interrupted, the card holds recent video.

At the core of the system is a HomePlug network device that plugs into the wall (for power and to capture the video signals). It also plugs into your router, so you'll want to locate this near your network interface. It's pretty much plug and forget, so where you locate the master controller isn't strictly relevant.

The system is controlled by either a Windows or Mac application which is used for setting up, configuring, and watching the cameras. There is also a Web application and both an iOS and Android app that lets you view your cameras from anywhere on the Internet.

My first test

For my first test, I had to unpack everything, install the wiring (it's a unscrew-and-plug-in affair), find six free wall sockets at far corners of my house, and then connect the master controller to my router.

I didn't pay any real attention to where the cameras pointed. All I wanted to do was make sure I could actually talk to them over the building power, and that the basic concept worked.

So, I plugged it all in, saw that the cameras each had indicator lights that were turned on, powered up the master controller software on my Windows 8 machine, and ... nothing ... for about two minutes.

The control software had grid squares for each of the six cameras and all six squares were empty. Then after a few minutes, one-by-one, they all just showed up. Each blinked in. All the cameras had connected and were sending video.

Granted, everything was upside down and a bit wonky, but that wasn't the point. The cameras were able to send video over the internal power lines, and — in what was a completely pleasant surprise — it all just turned on and worked.

The only odd thing was that each camera was sending a black and white signal. After a few minutes, I found the update control screen, and — holding my breath — told the system to update the cameras. Each camera obediently updated its firmware, and I suddenly had color, pretty-much-HD video coming from six zones inside my house.

I sent Jerry the electrician a note and scheduled a time. Still doubtful, we made an appointment for him to come over, but he ended with "I'll believe it when I see it."

Stay tuned. Tomorrow, I'll review the product in depth. Does it hold up to my initial positive impressions? You'll have to wait and see.

Here's a promo video about the product from Logitech. Wait until you read my review before deciding if this is for you or not. The product has both strengths and weaknesses.

Security is an area of considerable importance to me. While most of my work involves cybersecurity issues, I have long contended that homeland security begins at home. Since I work from home, and some of my work involves sensitive topics, building a secured home office has always been a high priority.

My wife vetoed plasma turrets, flame throwers, and automatic paint ball machine guns. The jury is still out on the motion-sensing pepper-spray yard sprinklers.

Shortly after buying this house, we started with an alarm system, a reporting and patrol protocol involving the local police department, motion-activated floodlights, and a perimeter security solution.

Okay, to be fair, I probably don't need all that, but I'm a geek and security systems are cool. So, well...you know...gadgets. Besides, this is the first time I've actually owned my own house and have the freedom to install stuff without having to apologize to my landlord (explaining the T-1 line running over my bathroom mirror was interesting) or ask permission.

I also had some difficulty with active defenses

My wife vetoed plasma turrets (too much permitting paperwork), flame throwers (apparently, setting fire to the neighboring woods is a negative), and automatic paint ball machine guns (paint splatters on the fence is also a negative). The town vetoed my idea for sod-covered, 12-foot deep trenches, insisting that town ordinances require deadly traps to be set back at least 25 feet from public areas. The jury is still out on whether or not we can install motion-sensing pepper-spray yard sprinklers (the town is still trying to figure out which permit is most applicable).

Fortunately, guard dogs are allowed, although personal, dog-sized plasma guns and projectile blasters were vetoed both by my wife and the city planning office. So far, no one has vetoed my use of a Parrot Drone, controllable from my iPhone, but I also haven't come up with a payload that's both an active deterrent and light enough for the drone to carry aloft for an extended flight.

All of that brought me back to video surveillance.

It isn't quite as cool as keeping a large bear in our back yard, but video cameras also eat a lot less. Once I accepted the fact that my wife doesn't really want armed guards patrolling the premises (except for celebrating meaningful special occasions, like my birthday, election day, Super Tuesday, and Black Friday), I decided to add video surveillance.

None of this is to say my wife is even the slightest bit unreasonable about important security items (or holiday celebrations). In Florida, it often rains on Christmas, so we've already reserved our Marines with umbrellas.

Unfortunately, most of the solutions I found (for video surveillance, not rain protection) seemed to have been designed back in the days before Windows XP came out. They knew virtually nothing about the Internet.

Most video surveillance systems of any note involve their own dedicated DVR (digital video recorder). They're self-contained systems that record activity to a DVR, and if you want to view that activity, you actually watch it on a monitor tied directly to the DVR.

Talk about old school!

I wanted a system that could be viewed, in real-time, from anywhere on the planet (including any room in the house). I wanted a system that could record video and upload that video to one or more secure repositories online. For this, the DVR solution wasn't appropriate.

Basic webcams wouldn't work either, because they're not really robust enough to survive outside reliably. I did, however, find some IP-enabled surveillance cameras, and some basic webcam security software, and tried that out. I eventually went with a different solution, and I'll tell you about that in the next article.

For now, let's talk about the solution I didn't go with.

My specs for cameras were pretty specific. I wanted both day and night vision. I wanted secure enough containment that normal (i.e., non-hurricane) weather couldn't damage the gear. And I wanted the camera to move, so I could point the camera at any location and see what was there.

I settled on the LOFTEK Sentinel D2. It had a lot going for it — and a lot of problems.

The biggest problem was that mounting it was a royal pain. It wouldn't work properly if mounted to a soffit (homeowners know the word "soffit" is the roof overhang — I never heard the term before owning a house). Because the mounting bracket was inflexible, mounting it to the soffit meant that the camera would look up (to the sky) and down (to the ground), but wouldn't track really well looking out over the yard.

Mounting to the wall of the house would be a pain, too. To begin with, I didn't want to cut into the finish of the house. It also required an extended mounting box (not included), because there are a ton of wires that come out from the thing, and they needed to be housed near the camera (because the connectors aren't all that long).

Worse, it had a ring of IR LEDs, but they shot straight down, so while you'd get night vision, you'd mostly get clear, infrared images of the tops of things. It just wasn't practical.

The LOFTEK Sentinel D2 had both a WiFi and wired Ethernet connection. I have pretty solid WiFi extending outside of the house, and with good wireless security that wasn't much of an issue. For network security reasons, I didn't want to extend even a single connector of my wired network outside of the walls of the house, so I wasn't going to run Ethernet to outside cameras.

The camera was usable, but I wasn't happy with it. The software interface was "meh," and while there are third-party software interfaces, they were generally unimpressive as well.

Even so, for a while the LOFTEK Sentinel seemed like about the best of the outdoor, IP-based solutions out there. Then, in one my many searches, I stumbled across the Logitech Alert system, which proved to be a win. I'll tell you all about that in the next article.

I'm actually quite glad I didn't go with the LOFTEK solution. Barely six months after purchasing the camera, it has vanished off the company's web site.

There is a stank in my local AT&T store. The place quite literally smells like damp, dirty socks.

For years now, my colleague Jason Perlow and I have gone back and forth over the question of whether retail is dead or dying. He contends it is, and I contend it's a business model issue, that some stores are thriving while others are dying.

In the spirit of full disclosure, I have an admission to make. I go into actual retail stores very rarely. Very, very rarely. I never liked shopping, I have an Amazon Prime account that meets nearly all my needs, and my work schedule keeps me pretty busy.

Besides, my wife actually enjoys shopping, so she does almost all the local errands. This isn't a gender thing. My dad loves to shop. I just never got that gene from him.

All that brings me back to the AT&T store. I needed a phone case from the store, and it was more convenient to pick it up than wait for Amazon to deliver it to me on Monday.

There was a problem. My wife wouldn't add it to her errands. She refuses to go in there. She says the attitudes of guys who work there are intolerably chauvinistic, in that "you don't know what you're doing, little lady" kind of way.

We both have iPhones on AT&T, so she's had a bunch of occasions to go in there. I bought my phone online, but she got hers in the store. She's also been back in there to look at phone cases and to get unnecessary bandwidth charges removed from her bill (she wasn't the only one).

So, fine, I'll run my own errand. It's only fair. I had tried to call first, but given that they sell phones, I found it a little disturbing that they don't ever answer their own phone when it rings. I called. No answer.

I drove the seven minutes to my local West Melbourne AT&T store. As I got out of the car, two women were getting into their car in the next parking space over. One woman turned to the other and said, "I hate that guy," pointing to the manager exiting the store. She continued, "He's always such a condescending jerk when I come in here."

This did not bode well. Not only was my wife distinctly unthrilled with this store, so was at least one other woman. Ah, well. My needs were simple. I went on in.

The very first thing I noticed was the stank. It smelled like a gym, on a particularly hot day. To be fair, I live in Florida, but it was only in the mid-70s yesterday. It just wasn't hot enough to justify the smell of a locker room.

The store consists of two main counters on the left and right side of the store. There were two salesmen behind the left counter (nearer the entrance) and three more behind the right counter. Each was with a customer. In the middle of the store were free-standing display areas, mostly uninteresting.

Some phones were on the display kiosks and were actually powered up and working, unlike the computers in our local office store (see Have PC retailers lost the will to live?).

That retail experience took me 30 minutes. By contrast, I could have ordered what I needed from Amazon in 30 seconds.

There was a definite Glengarry Glen Ross feel to the store. The man closest to where I planted myself was explaining a contract. The woman to my left was spending a few hundred bucks to buy a MicroCell so that the cell phone she was already paying for would actually work at home.

There were no female workers in the store. While I didn't see any outrageously inappropriate behavior, the sales guy who finished up first looked at me, and the only thing he said was "I'm going home." The store is listed closing at 8pm, and I was in there at 7:30, so he was clearly in a bit of a hurry to get out of there.

He didn't bother to say "Hi". He tried to avoid eye contact. He didn't offer to introduce me to another sales guy. He just wanted out and didn't care about his store's reputation, the possible purchase I might have made (at that point, he didn't know if I needed a phone case or an S4), or the impression he'd make on someone who might be a repeat customer.

He just wanted out.

In any case, it only took me about 15 minutes to get my needs met. I'm hard to miss, so the next free sales guy came over to me as soon as he was done with his customer. He was polite and got me what I needed. I had no complaints.

Even so, as I walked out, I thought once again about Jason's arguments about retail. It took me seven minutes to drive there and seven minutes home. Call that fifteen minutes. It took fifteen minutes of standing around. I didn't like how I was treated. There were other shoppers who clearly didn't like how they'd been treated.

And I had to get there before 8pm. And then there was the stank.

That retail experience took me 30 minutes. By contrast, I could have ordered what I needed from Amazon in 30 seconds. Had I been willing to wait the weekend for delivery, I would have saved half an hour.

Now, to be clear, half an hour isn't that much of a time investment. But when the choice for us all is sitting at our desks or on our couches and hitting One-Click or driving 30 minutes, 40 minutes, an hour, or more and having to endure waits, standing around, inattentiveness, attitude, or the stank, more and more of us will choose the online experience.

My only disagreement with Jason is that I don't believe retail is destined to die just because of "the online". If retail is dying, it's because the retailers themselves are letting it happen.

This store is simply poorly managed. There was no reason a salesperson should be allowed to let a customer feel ignored. And, of course, there was no excuse for the stank.

But Jason is right. If retailers don't change their ways, retail will be a wasteland. And retailers will have nobody to blame but themselves.

It's not consumer behavior, and it's not Amazon's fault. It's bad management. And, of course, it's the smell of failure.

Update: I've been contacted by a number of officials at AT&T who have expressed concern about this report and have told me they are looking into it.

Gretchen Schultz, an AT&T spokesperson, provided a response for publication: "Our goal is to delight our customers when they enter an AT&T retail store. Most of the time we get it right. Sometimes, regrettably, we fail to meet expectations and when we do, we value the feedback our customers give us and strive to improve."

I really dislike software updates. It didn't used to be that way. It used to be that I looked forward to the new features, new capabilities, new toys.

But that was before I had a life and responsibilities. Now I have both, and I find myself to be somewhat more change averse.

Today, a day I had a lot of other more important things to do, Google decided to break one of my most-relied-upon resources: Google Chat.

Let me be clear: I use Google Chat for work. I talk to many of my colleagues about work-related activities, about ZDNet editorial, and about projects I'm working on.

I don't hang out.

I don't use Google Chat to make new friends or to share small details about the danish I had for breakfast or hear about new TV shows. I use it to coordinate meeting schedules, deliverables, titles, abstracts, article drop times, and the like.

My Google Chat list is carefully curated. There are about 30 people on it, all of whom have something to do with work. Robert Scoble, as much as I like Robert, has never been on my Google Chat list, because I don't work with him.

Today, all of that changed. My Chat list is gone. In fact, my nice little Google Chrome extension for Chat is gone, as well. Poof. Stolen, like the time stolen from me today trying to recover my work messaging system.

My Google Chat list is carefully curated. There are about 30 people on it, all of whom have something to do with work.

In its place is Hangouts (and Robert Scoble is on the list). I had a British professor who used to titter every time someone mentioned hanging out, because to him, hanging out meant, well, something very inadvisable in public.

I don't want to hang out (either using my professor's definition or the more mundane one) with my work associates. I want to send a quick chat and get back to work. So do they.

This morning, however, I have to figure out what happened to my list of work associates, figure out how to get Chat back, figure out how to stop everyone in my Google Plus (which today is such a minus!) from hanging out on my desktop, and then wonder if every other work associate I've got has the same issue, and whether all of our productivity for the next week has just gone down the chute.

So, thanks Google. Thank you so plus'n much.

I recently read an article on Wired.com entitled, Why E-Mail Newsletters Won't Die. Like most Wired pieces, it was well-written and thoughtful.

Unfortunately, while the author extolled the renewed virtues of email mailing lists, he didn't really cover some of the modern-day realities and challenges most marketers and list managers won't tell you about until it's just too late.

But I will. I've been there. I've done that. I have the scars to prove it.

The premise of the article is simple. Despite the prevalence of social networking, the rise of Twitter and Facebook, nothing works for direct, immediate outreach in the corporate space like an email newsletter.

This is true.

We have a number of email newsletters here at ZDNet, including one devoted to my ZDNet Government blog. You can see a big, blue subscription form. It's right there, to the right of this column ------>.

Back in 1998, when I first started the ZATZ Web sites, we knew that people would come visit the Web sites (even then, search mattered). But we knew what would bring readers back was reminding them we were there. So every week, we sent our opt-in readers news summaries and tips and techniques.

Over the years, we amassed a list of more than a million subscribers. As email spam became more and more of a problem, we added further insurance to make sure only readers who wanted email got it. We added a triple-opt-in process, which meant you had to opt-in, then you'd get an email confirmation where you had to confirm the opt-in, and then we had a permablock option, where you could permanently opt-out of getting any email from any newsletter, forever.

Further, each email message sent had what's called a VERP, a variable envelope return path, that allowed us to be sure that regardless of what email address the user used to subscribe, when he or she wanted to unsubscribe, we could guarantee we could successfully process the unsubscribe request.

I decided I'd try to move my mailing list to a list management company. The process was a total disaster.

In addition, we would periodically validate and de-bounce the list. Validation meant sending the list to outside providers to verify that each email address was deliverable (and removing the bad addresses) and de-bouncing meant that after a certain number of failed deliveries in our own system, we permanently removed the address.

In other words, we operated our mailing list with the absolute best practices. We'd get one or two subscriber complaints a year, usually from someone who'd forgotten he or she had subscribed.

A year or so ago, I decided I didn't want to run my own mail servers anymore. I've been fortunate enough to change my career primarily to teaching, lecturing, writing, and punditry. Mailing list management is a pain. So I decided I'd try to move my mailing list to a list management company. There are a ton of cloud-based mailing list providers out there, most with very inviting offerings.

The process was a total disaster. Even though I had a thoroughly vetted set of lists with subscribers who opted in, it became impossible to actually move the list to a new vendor and keep it running.

Basically, what happened is I ported the list to a new vendor, and did a first mailing. Since this first mailing came from a new source, some of our subscribers reacted by complaining they'd never subscribed. Even though they'd subscribed to our newsletters, they weren't aware of the new sending email address of the new vendor.

When you have hundreds of thousands of subscribers, you're going to get a few complaints when you do something new. A few complaints came into the first mailing list provider, and without even contacting us, they just cut us off. No mailings and no access to our list. No calls returned. No email contacts returned.

Now that I knew this might be a problem, I moved on to another vendor. This time I did my homework again, and added relationship building to the process. I informed the vendor ahead of time of the situation, discussed it in detail, worked out exactly the potential issues, and had complete assurance that there would be no problem moving the list over.

We started our first mailing, and wouldn't you know it? A few subscribers complained. The second vendor immediately cut me off as well. What annoyed me the most was that there was no acknowledgment of our previous discussions, and absolutely no willingness to work out a solution.

To be clear, the newsletters I mailed were not spam. One was about IBM's Notes and Domino, another was about mobile technology, and a third was tips and techniques for Outlook and Exchange. All three had been successfully mailed for more than a decade.

Vendors are so scared of being slapped with the "spam" label that they don't use good business practices when working with customers.

I actually tried a third vendor, but that one insisted that they wouldn't mail to any pre-existing addresses. We'd have to start from scratch, and they'd be willing to build up a list, but they wouldn't take anyone who hadn't opted in on their mailing list system. Period.

Finally, I just went back to operating my own servers, using the nice ISP I've been using for years. Because the ZATZ sites are now primarily archive sites for the hundred thousand or so articles we've published over the years, I no longer offer a subscribe form. We just mail weekly updates to our existing, very loyal readers, who number now somewhere in the 20-30 thousand range.

So, what's the moral of this story?

It's sad, really. Vendors are so scared of being slapped with the "spam" label that they don't use good business practices when working with customers. It's a one-strike and you're out sort of game.

That leaves me with one warning for you: if you're currently operating a mailing list with a lot of subscribers, work your vendor relationship carefully. Because the odds are, you're stuck with your current solution. It's highly unlikely you'll be able to move your mailing list to another vendor (especially if you're a relatively small company with a relatively big list).

Email-based mailing lists may be here to stay, but so are the systems they're mailing from. This is lock-in and you're pretty much stuck with what you were doing back in 2005.

This is big. End-of-an-era big.

For most of the PC era, software was sold as packaged products. Before the internet (and, really, before nearly ubiquitous broadband), PC software (and Mac, too) was sold on disk — first on floppy, then CD, and eventually DVD.

In most cases, you didn't officially "own" the software. The "shrink wrap" agreement said that you were merely granted a license to use the software according to some very long, complex, and often incomprehensible terms, which often included things like the promise to make pancakes for the software maker every weekend for a decade and do laundry every other Wednesday.

Say "monthly fee" and imagine you're a product manager at Adobe or Microsoft. Doesn't it just make your toes curl up and make you breathe faster?

Once broadband came around, many software products that were sold in physical packages were also offered as downloads. Even so, the process was generally the same. You'd install the software, enter a license key, and be granted the right to use the application on one or sometimes two computers.

App stores changed the game a bit, but they still followed the "you buy it once" model. When you bought Angry Birds (OK, that's not a good example, because most of you have bought all the Angry Birds variations, on all their platforms). Let's try again.

When you bought a not-Angry Birds app, you bought it. You'd pay your $1 or $3, and it was yours to use. Sure, if you downloaded a game, you might get suckered into spending $100 on magic crystals, but that was your option (or obsession), not a requirement for continued use.

On the other side of the industry was the burgeoning new business model called software as a service (SaaS). Instead of buying a piece of software, you'd rent it — but you also didn't have to install it. It ran on the web.

Gmail is an SaaS product, a web application. As far back as 2005, I signed up for QuickBooks Online. I was moving my business a thousand miles from New Jersey to Florida, and the one thing I absolutely needed, even while all our office computers were in the moving truck, was our accounting data.

We signed up for QuickBooks Online, and in return for an ever gradually increasing monthly bill, we got access to our books anywhere, and I no longer had to maintain the accounting server. The online version was slower than the installed application version, so our bookkeeper regularly complained, but all I had to do was nod sympathetically. Because Intuit maintained the product, I didn't have to add an action item.

Since then, an entire, huge market of web applications have sprung up, many replacing the C:\Program Files and C:\Program Files (x86) installs we've all come to know and love so well.

Packaged software was being compressed on two sides. Well, actually, three. In addition to the app business and the SaaS business, those individual and small-fry developers that used to be called "shareware" developers simply continued their mode of distribution — which was to distribute their software online. The shareware folks used to limit their software or nag you incessantly, but really they were simply using online distribution — without all the cost of packaging.

So, once again, the big vendors — the Adobes and Microsofts of the world — were being pressured on three fronts. Mobile app stores were dragging down prices to ridiculously low levels, for relatively low-quality software. Small-fry shareware developers were distributing online without packaging cost. And SaaS vendors were selling entire suites of web-based products for a monthly fee.

Ooh. Did you feel that tingle? Say "monthly fee" and imagine you're a product manager at Adobe or Microsoft. Doesn't it just make your toes curl up and make you breathe faster?

Can you imagine? Instead of customers paying once every two or three or four years for a product, and then having to hawk upgrades to keep sales up, and instead of a measurable percentage of deadbeats cracking license generators and pirating the software, selling software as a service would mean a steady, constant, predictable income. Every. Single. Month.

The thing was, some applications couldn't really run well inside Java, JavaScript, Flash, HTML5, or whatever web-based tool you want to name. Some applications need the core horsepower of one or more local processor cores and compiled code. Photoshop is a good example. So is Excel and PowerPoint.

Sure, Google Docs implements a limited word processor and spreadsheet, but for all — all — the power of the Office applications, you need a local processing core and compiled code.

So, what Microsoft did was come out with Office 365. It's basically packaged software, sold as a monthly service. I signed up for it. Read My big email switch: Why I picked Office 365 over Google Apps.

For about $30 per month for my wife and I, we got our email and 10 licenses for Office. Since we each use three computers, it was somewhat less expensive than going out and buying individual copies of Office (and I need real PowerPoint for work, I can't use a simple slide-making clone).

But $30 per month is $360 per year, and if Office updates every two years or more (and, to be honest, two new features from PowerPoint 2010 to PowerPoint 2013 isn't really an upgrade), we'll really be spending quite a lot for software we once used to just buy, outright.

In the case of Office 365, the factor you might not know about is that we also get Microsoft's Exchange service, a service that we were already paying $20 per month for from another vendor. So, technically, we got the Office part of the puzzle for an extra $120 per year — not a half-bad deal.

Adobe just announced that it's taking this subscription model one step further. Adobe will no longer (after Creative Suite 6) sell stand-alone software. From now on, it's monthly fee or no software.

Before I go on, I need to make a disclosure. It's been a long time since I've bought Adobe's products. As the editor of Computing Unplugged, Connected Photographer, and host of DIY IT here on ZDNet, Adobe has provided me with copies of Adobe products since the first Creative Suite came out back in 2003 or thereabouts. Adobe was also kind enough to provide me with a subscription to Creative Cloud, so I can evaluate and show you how to get the most out of Adobe products for you here on DIY IT.

Even so, as a small business owner for so long, I can see how this change might be disturbing. While buying the full Master Collection was an expensive purchase, it was something a business owner or designer could do once, and then not buy again until cash flow would allow for it.

Of course, that was a big purchase. Amazon lists CS6 Master Collection for between $2,100 and $2,400, and even the student edition is nearly $1,000. That's not just a car payment. That's a crappy car.

That kind of cash is hard to come up with, where $40 to $50 per month is much easier. I imagine Adobe will tick off some of its more set-in-their-ways customers, but bring in a lot of people who want the full power of Adobe products, but who can't (or don't want to) come up with thousands of dollars.

More interesting, though, is the accounting question. When you bought a software product like Master Collection, you had to depreciate it over years. But if you buy a monthly service like Creative Cloud, you can expense that on a monthly basis.

So, some people will derive a benefit from the software-packages-as-a-service model, and others won't be happy about a new monthly bill.

The thing that's a little interesting and a little disturbing is how all these new monthly bills add up. When you add up the email application, the graphics design suite, the office suite, the cloud storage, the accounting web application, the group video chat application, and the desktop sharing service, each Windows and Mac desktop user will probably find themselves saddled with what amounts to another car payment.

So, here's the question: Do you really need another car payment? Do you like this software-package-as-a-service model (and, remember, most require you to sign up for a year at minimum), or do you wish we'd return to the days when we bought what we bought and vendors stayed out of our wallets until we let them back in?

Comment below.

Microsoft needs a next-generation product, because the first generation product hasn't been much of a barn-burner.

Given the existing Surface device's relatively poor success in gaining traction in the market, it might be useful for us to look at five features that — taken together — could raise the product's profile in a second-gen product.

7-inch form factor

The CNET article reported that Microsoft is readying a smaller device. That would be a relatively big step for Windows-based machines (especially desktop-oriented Windows machines), since there are relatively few that run on small displays.

This could actually be a big win for Windows 8 users, because there are many uses of Windows that could benefit from what would essentially be a small, smart flat-panel display.

Beyond the full Windows 8 experience, even the anaemic Windows RT could benefit from a smaller device. Reports are that Apple's iPad mini is outselling the full-sized iPad. I have both an iPad mini and an iPad, I very much prefer the mini because it seems to fit into my work style more conveniently.

And as ZDNet Health blogger Denise Amrich (our resident RN) reported in an interview with two health professionals, a tablet-based Windows 8 may have some unique advantages in healthcare.

The Xbox Arcade library

In The Verge's coverage of a possible new Surface, there's mention of an Xbox Surface. Way back in June of last year, I wildly speculated that Microsoft would announce an Xbox tablet.

That obviously never happened, and it's somewhat unlikely that it will this June either. We'll know more on May 21st, when Microsoft is widely expected to announce the next generation of Xbox, what everyone is calling Xbox 720.

The current-generation Xbox 360 is based on a 64-bit PowerPC architecture, while the original Xbox was based on a Pentium III architecture. Obviously, with the current Surface devices running ARM and Core i5, getting Xbox applications to run on the Surface might be a challenge.

That said, Apple has a history of emulating PowerPC on it's older Intel machines (using the now-discontinued Rosetta emulator), so it's possible to do.

I see two possible courses here: First, while it would be impractical to run the full-experience Xbox console games on a new Surface without a lot of additional hardware (and heat-dissipation improvements), a new Surface could run Xbox Arcade games.

The Xbox Arcade is hugely popular, and might prove to be a very compelling reason for consumers to pick up a Surface. It would almost immediately counter Android and iPad's game-availability advantage.

The second possibility is that Microsoft may announce a move off the PowerPC architecture for Xbox 720, and there may be an opening for lower-powered games to run on Atom or Core i5 in a refreshed Surface. Frankly, I don't think that's the barn-burner move. I think introducing Arcade on the Surface would make for a market winner.

Lighter and more comfortable to hold

One of the failings of the Surface is that it's great on a desk, but it's not quite that comfortable to hold. If ever there was a critical design flaw in a tablet, that's your problem, right there.

ZDNet's James Kendrick reported after 45-days of using his HP Envy x2 (a full Windows 8-based convertable tablet), the tablet part of the Envy is lighter and easier to hold than an iPad — and it's running full Windows 8.

If HP can do it, so can Microsoft. In fact, Microsoft must improve the hold-a-bility of its tablet or it stands almost no chance of widespread acceptance.

Price price price

Speaking of widespread acceptance, there are two relatively hot competitors to the iPad — the Nexus 7 and the Kindle Fire. Why? The answer is really quite simple: Price.

These are two great devices that were released for $200, a breakthrough price for a tablet. Consumers are very price-sensitive, and if they're going to adopt something widely, Microsoft will need to offer a much less expensive offering or a next-gen Surface just won't stand much chance against its more entrenched (and less expensive) rivals.

Compelling Metro apps

I have asked Windows 8 user after Windows 8 user (and the few Surface users I could find) about what good Metro (uh, Modern UI) apps there are out there. The universal answer: None.

I'm writing this on Windows 8 right now, and I have access to the entire Modern UI Windows store. I tried to find a good text editor, but they all sucked. Not just mediocre, but terrible. So I'm happily using my traditional Windows desktop applications — fine for desktop use, but not a compelling tablet app solution.

Evernote makes nice little tablet apps on the iPad and even on the iPhone, but the Metro app for Windows 8 is pretty much unusable.

I know there are some useful Metro apps being released for verticals, and vertical-market custom software has always been a strength of Microsoft's, but unless Microsoft builds up a stable of really good, best-of-show Modern UI apps, the entire tablet/Start screen/touch model of the Surface will go the way of Microsoft Bob.

But if Microsoft does follow these five pieces of advice, they'll have a winner. And Bob's your uncle.

Tiobe Software maintains what it calls its TIOBE Programming Community Index, which cross-references language popularity among the professional programming community.

Objective-C, which is the language most used for iOS (and Mac) development, has been skyrocketing since the original iPhone App Store opened. Back in 2011, there was a little over 1 percent drop (among all programmers and all languages). But since around December, Objective-C popularity has dropped by more than 1.5 percent (among all programmers and all languages).

It peaked in December at 11.116 percent, but has now dropped to around 9.5 percent. Off the Apple platforms, C, Java, and C++ still hold the top slots, respectively.

I understand. I miss the Palm OS. However, I have moved on. A long time ago, I moved my notes and contacts to Outlook, and use an iPhone as my daily driver.

However, there are still some of you out there, like reader John. John told me:

I join you in the secret Palm handshake. I pre-ordered (!) the first every Palm. We just knew it would be good (but I did own a Newton also, sigh). I still have the earliest Palm model in the basement. I have a Zire 72s still running strong. To heck with the hardware, though, it is the desktop app that makes my day. I have 1,000+ entries in my contacts as I used it to store every kind of information. I haven't found a reasonable application to take its place. But running Palm Desktop on Win 7? No issues, man. The only issue is getting sync to work. Luckily, Acceca came to the rescue with the necessary 64-bit drivers. Free! Cool!

And so, that's what I'm here to tell you. There are apparently 64-bit drivers for you Palmists who want to sync your devices on computers made in the modern era.

Here's the link: 64 bit USB Windows Driver for Palm/Garnet OS.

I love it when readers help out like this. And as a result, John gets our second official ZDNet DIY-IT Helpie McHelpypants Acknowledgement. Nice job, John!

Finally, the entire PalmPower and PalmPower Enterprise Edition archive (from back in the Palm heyday days) is still online and searchable. It's now part of the archive I run at Computing Unplugged. Disclosure: I get a few bucks every so often for ads that still run on the site. In any case, there are more than 20,000 articles up there, going back to 1998.

Related stories

• Why old people still like their PDAs

• Did you know you can use Palm Desktop with your Android phone?

Once you get past the idea that the cloud isn't a place full of soft, fuzzy foam, but is a wide array of really huge datacenters with thousands of servers and a power load to rival a small city, cloud concepts start confusing people.

Nothing, it seems, is more confusing than the difference between public and private clouds, and beyond that, the safety and security implications of using either type of service. So in this article, I'll run through some of the basics.

The cloud datacenter environment

Let's first take a second and describe what a cloud datacenter environment is. It's a datacenter with a bunch of servers all running services (like email, CRM, ERP, etc). These services can be accessed via the internet, and they can be both instantly deployed and metered. In other words, a new user can be spun up quickly, and the cloud operator can charge that user based on how much he or she uses the service.

Many cloud providers also provide their customers access to compute power with very granular scalability. In other words, if you're buying cloud services, you don't have to buy a new server, rack, or datacenter to add users to the service. You just add accounts, user by user, resource by resource, and pay for just how much you use. This is great when you don't want a huge amount of capital investment to be tied up in compute-based physical gear.

So to summarize: Instantly deployed, metered, scalable, available online.

Public vs. private cloud

A public cloud is this sort of service, offered to everyone. Gmail is a public cloud. Office 365 is a public cloud. Amazon Web Services (AWS) is a series of public cloud services. Dropbox is a public cloud service. Why? Because anyone can use these services, as long as they pay for what they use.

A private cloud is pretty much exactly this same sort of service, except you own it and it runs in your datacenter. Instead of offering instantly deployed, metered services to the world at large, you offer these services to your internal customers, the various departments and organizations in your company.

The big difference, of course, between a public and private cloud is that if you're running a private cloud, you're physically managing the machines, paying for them, and operating all the infrastructure, including the software that does the instant deployment and billing of services.

Security implications

Given what you now know, it's pretty obvious that there's at least one major security difference between private and public clouds. With private clouds, you control the physical servers and access to the servers. With a public cloud not only do you not control the machines or access to them, you're unlikely to ever touch one physically.

As an end-user, if you store your data in the cloud, you're trading off physical security for data security. For example, if you lose your thumb drive or a laptop, your data is still secured in the cloud. But since your data is in the cloud, you're pretty much one exposed ID and password away from datamageddon.

From an enterprise point of view, there are some security benefits to a private cloud. Your information lives behind your firewall (unless you've co-located your servers somewhere else, and even then, you can add some firewall protection). Here are some other benefits:

• Your data also can live behind your own locked doors

• You don't have to connect to the internet and can completely isolate your data infrastructure

• You know exactly where your data lives

• You design the architecture for your exact needs

• You know exactly who is granted physical access

• There is absolute clarity of ownership

• There is no risk if your cloud provider shuts down

On the other hand, there are some disadvantages as well:

• Your employees have physical access

• You are on your own when defending attacks

• You are subject to the whims of nature

• You are subject to the whims of your ISP

• You are subject to the whims of your local power grid

• Your security is entirely your responsibility.

Now let's contrast that with the security benefits of keeping your data in a public cloud:

• Your data lives behind an enterprise-class firewall

• Your data lives in a very secure facility, often with multiple degrees of physical security

• Thieves intent on stealing your data may not know where your data lives

• Your gear is not at risk from disgruntled employees

• You gain security expertise from your vendor

• You are not alone when defending against DDoS

• You are protected from hardware failures

• You are protected from sudden surges in demand.

But as we've discussed, there are also some security disadvantages of using a public cloud. These include:

• Access can be granted from anywhere

• Your data must travel "in the wild" over the open internet to your cloud provider

• Your vendor might grant physical site access to other tenants

• You may be subject to jurisdictional issues, especially when you're dealing with international issues

• There is very little established case law

• You are dependent on the responsiveness of vendor

• You are dependent on the whims or quality of vendor.

It's the last one that can be a big concern. If you agree to one set of terms of service, but the vendor suddenly changes those terms of service, you could find yourself almost instantly cut off from your data and your customers.

I've heard a whole bunch of horror stories about agreements that companies had with their cloud vendors, and then suddenly, because of one minor problem (or even a database script run amok), they found themselves completely cut off from their data or completely shut down. Given that many cloud vendors aren't reachable by phone, it's entirely possible that the silver lining in cloud services could suddenly become a black hole.

You should also investigate how secure your vendor is. Is this a new, venture-funded startup that's one bad quarter away from closing doors? Or is this a company with deep resources that will clearly be around for the long haul? Does the company store your data across multiple datacenters, in multiple locations, and what sort of backup and recovery strategy do they offer?

What I do

Back when I started ZATZ, there really wasn't a "cloud". You had to buy your own servers and your own connectivity. We started with an ISDN line, then went T1 around 1999. I ran an entire rack of servers, and at one time, had to convince Verizon to put a T1 line into my apartment.

We were feeding bandwidth through my bedroom, over my bathroom mirror, down the hall, and into a former linen closet. I was told that we consumed the same bandwidth as the local college campus, and it took a few weeks of explaining to convince them that I really did need to feed a few million page views a month from a series of hand-built Linux boxes in my apartment.

However, when I got married and moved to Florida, I realized that Florida weather could knock our servers back to the Stone Age on a moment's notice. So I contracted with a co-location provider in Illinois (who operates out of a former nuclear bunker), and put four machines there.

That approach continues to work well. But as I've personally moved away from having to manage a publishing company, I've wanted to let others do more and more of the work. It started with moving our bookkeeping off the local share to Quickbooks in the cloud (that was 2005 or so), and has recently ended up with moving our email to Office 365.

My strategy now is to offload everything I can to the cloud that's not either bandwidth constrained (my media asset library needs to be fast, so it has to be managed inside the firewall on our tank) or based on a unique solution, like the CMS I wrote and that runs all the ZATZ archives.

Final thoughts

There's a lot more to cloud-based security, but this brief introduction should at least get you started.

Now if only I could convince my doctor to stop trying to get free advice from me while making me cool my heels for two hours in his waiting room because, while he has a cloud-based scheduling tool, he never actually pays any attention to it.

I'm talking especially about the makers of specialty PC software, particularly those who make what we used to call "utilities" or "power tools." In a world where PC sales drop precipitously, and our operating system makers seem to be dumbing everything down to an Angry Birds least common denominator, makers of deep, rich power tools are having some dog days.

These are the makers of the incredibly feature-rich text editor, developers of the screen capture program with 400 feature, authors of the thumbnail viewer that's really a full digital asset manager, creators of the file copy program that has more features buried in its graphical UI than the Linux shell has in all its arcane commands, and coders of the development environment that can do the craziest sorts of cross-platform live debugging.

Many of these vendors have been in business for a decade or more. They've been making money on one main piece of software and have continued to refine it, improve it, add customer-requested features, and chugged along, providing a unique value to a select set of customers with unique needs.

You get the idea. Power tool software for power users.

These companies don't sell their products to everyone. Here's an analogy. Most homeowners own a power drill. It might not get used much, but once in a while you might break it out. But few of us homeowners own a full drill press. After all, that's more like something you'd see in a shop rather than the typical home.

But some homeowners have home shops, and use drill presses, milling machines, and all sorts of customized tools. There's a market for these power tools. It's not just the same market that sells Hello Kitty-powered screwdrivers.

Over the past month or so, I've been working on a big image workflow project, and so I've been looking at and discarding an entire array of power tool software. I go through these phases when I put on my power user hat, and try to optimize a solution for some sort of unique work problem or process.

During these phases, I often sift through a couple of dozen tools, looking at the problem from a wide variety of angles, and get to see some very slick, deep, special-purpose software.

The PC software industry also goes through its phases. It's had its ups and downs for decades now. There are times when there seems to be no end of opportunity and times (like now) when things seem particularly bleak and sales are down.

When the market is down, being a software company owner can be a rough, rough gig. There are often not quite enough new customers to support the workload, so something has to give. Often, it's the support staff that's the first to go, followed by good nights' sleep, and perspective.

If you're an owner/developer, you're struggling with trying to balance development time with support time with the time you devote to any other forms of income that are being used to prop up the business and pay bills.

If you're an owner, but reliant on a developer either as a partner or a contractor, things get even more worrisome, because somehow you have to keep the developer in pizza and caffeine, and keep him or her happy enough (and paid enough) so you can continue to keep your product evolving and on the market.

I know. I've been there. I ran a company that made and sold specialty software (in my case, plugins and embedded database technology) for over a decade, before I finally sold off my flagship product. I then wrote The Flexible Enterprise (free PDF download), about how to transform a business model and create an agile business.

Before then, I was Symantec's director of product marketing. I've been on the board of the Software Entrepreneur's Forum (now SDForum), and in my subsequent decade as a publisher of online technology magazines, worked with hundreds of other software vendors. I even wrote 40 rather silly iPhone apps and made a few bucks off that experience.

In other words: been there, done that. I think I'm reasonably qualified to give advice to the struggling PC software vendors out there. Now, I'm fully aware your situation is different: you have a plan, you know what you're doing, you don't have time, you know you should, but... look, I've given all those excuses. Like I said, been there, done that.

But the bottom line is simple. If your business is struggling, more and more of the same ol', same ol' isn't going to suddenly give you a lift. You need to make a few changes — and, fortunately, they're not all that hard or costly — or the vicious cycle will continue.

Roll up your sleeves. Let's get started...

Update and modernize your web site

If there's one universal thing I've noticed in this power tool search, it's the really sorry state of web sites run by PC software vendors. Oh. My. Gosh.

These sites often look like they haven't been updated since 2002 or 2003, like they made a big push to get a product out for XP, but haven't really done much since then. It's clear products have been updated, because the release notes constantly show improvements and point releases every few months, but often the last "news" item in the news section is from 2008.

Often, the online documentation doesn't match the latest version number. The manual says it's for version 11 of the product, but the release notes indicate that version 15 is now out.

If there are video tutorials, they were often created pre-YouTube and are downloadable, low-res files, rather than embedded YouTube videos.

It's terrible. Despite all the work you put in each day holding things together, and the quiet desperation you live with in the dark of the night, it looks to all the world like you don't care about your products and your business. New customers who visit your site will click away almost instantly, because your site looks like one of those old, untouched, ghost town sites that live all over the Internet.

So here's my advice: if you're going to sell software in the Internet-age, you must (must, must, must -- more than anything else -- must) modernize your Web site.

This isn't hard to do. There are thousands of quite fine WordPress templates, hosting providers who can do complete Web site installs for a click of a button, plans that won't cost you more than ten bucks a month, and so forth. If you don't have the graphics design skill, use one of the pre-made templates.

In fact, I'll help you here as well. Here are some of the vendors I've used:

• iThemes
• Elegant Themes
• WPMUDev
• ThemeForest
• PageLines

I know this process could take a week or so you don't have, but trust me on this: if you don't do this soon, you'll have all the free weeks you want.

Deal with your forums

You know there's a problem when forums are populated with "Hello, is anyone here?" messages. It's particularly bad when the main web pages point customers to the forums for support, and the forums are populated with complaints that the company doesn't answer or brutal complaints about products, product strategy, updates, etc.

Here's a hint: if you can click into your forums and before you even scroll down, you can see a pile of "this company sucks" messages, you need to deal with your forums.

My advice (and this is my software company owner hat, not my transparency journalism hat) is to shut down those forums. Just kill them. I know there's a lot of knowledge trapped in there, but there's also a lot of bile you just don't want new customers to see.

Besides, your forum software probably also looks like it came from the days when Bill Clinton was President. It, too, probably needs a big face lift.

Many of the theme vendors I've suggested above also offer forum software, and there's also a ton of free downloads. Heck, phpBB might be annoying, but it's free and looks at least like it came from a year with a "2" in it.

My other forum tip is this: check over your forums at least once a week. If you can't scan your forums, find someone (even your kid, your mom, that annoying but very loyal customer ... someone) who can look over the site and let you know when strongly negative posts need to be dealt with.

If you can, solve the problem or make good, don't just sweep those problems under the carpet. But no matter what, if you want to keep selling software, you can't have that sort of activity on your forums. It's the kiss of death.

Scan for old dates

While we're talking about the kiss of death, another sure-fire way to scare away new, prospective customers before you even know they came-and-went is to have lots of old dates on your site.

Seriously, if most dates on your site show 2009 or earlier, you're in trouble. It's okay to have release notes that end in 2013 and start back in 1999, but other than that, you really should make sure no date on your site is older than 2012 -- especially if its your software product.

I found one product in my search for a good media asset manager that listed its last update as 2005. Given the importance (to me, anyway) of my project, I couldn't take a chance on something that hadn't been updated since before Vista.

Keep reading. Tough love is good for you...

Update your installers

Another easy thing to do is update your installers. Even if the software you're installing is pretty much the same as you've been selling and updating for years, that installer that is sized for eight-character file names and likes to stick things in the root C:\ directory is going to seem completely anachronistic.

There are some great, free installers out there. I know it's another weekend's worth of work, but you do want to keep your company, don't you?

Update your system requirements specs

If the last OS listed in your system requirements is Vista, you have a problem. If you list NT in your system requirements, you have a bigger problem.

Test your product against Windows 7 and Windows 8, and then update your system requirements to explicitly call out that the product works on Windows 8. Don't worry about the idiotic "Metro" interface, but if you say you support the Windows 8 desktop (and you do), then your prospective customers won't think you've been in suspended animation since the Bush administration.

Post a frickin' phone number

Yeah, I know we're moving to a world where all interaction is online. But customers who want to buy stuff sometimes want to call and get an answer to a question right now.

I also know that posting a phone number is an invitation to your current customers to call, bitch you out, and ask you questions you can't answer, but that dialog is good, too. It'll reinforce to your customers that you're still there. Plus, you never know. That really loyal (if cranky customer) might just have a way to save your bacon -- simply because he needs your product as much as you do.

Acquire some other products

Most of the software tools vendors I've seen are one-trick ponies. They have their one product, and that's it. That means if you make a sale to a customer, you're done. There's no more money to be mined from that hole until you release a major update.

One great way to augment your income (and help maximize utilization of the business infrastructure you already have) is to acquire publishing rights for other programs. This isn't quite as easily done now as the days when distribution meant brick-and-mortar and manufacturing meant paying for packaging, but there are still a lot of talented programmers out there who don't want to do the administrivia.

Don't contract out to have something built. Instead, scour the Internet for great programs with poor descriptions, support, or web sites. Offer to publish those products, update the manual, sell them, and provide a royalty back of 15-30 percent after your cost.

You'll probably need a contract to do this, but don't go running to a lawyer. First, most lawyers don't know squat about the software business, and you'll wind up paying more for the contract than you'll make selling software. Instead, find another software vendor who isn't a direct competitor and ask them if they've got a publishing contract they don't mind you using.

Make friends with other software vendors. These sorts of mutual-support contacts are very helpful.

I know you may not think you can afford to acquire a product, but trust me, it can be done. In fact, I wrote two chapters on this phenomenon in two of my books. Read the first chapter in The Flexible Enterprise and read "Luck is a skill" in How To Save Jobs. Both are free downloads, so there's no excuse not to.

An even easier approach (although a little less profitable) is to resell other existing products on your web store and in mailings to your existing customers. Find complementary (or even competing products) that would appeal to your class of customer, do a deal with the software vendor to resell it (you'll usually get 40 to 60 percent of the selling price).

Do yourself a favor and start this process now. An extra five or ten SKUs will do wonders for your incremental income.

Know when to hold 'em, know when to fold 'em

Finally, we're at the tough love stage of our discussion. Sometimes, it's just time.

Look, if your product really hasn't been updated since 2005, if you can only run on 32-bit XP, and you crash constantly on 64-bit systems or anything running Windows 7 or later, it may be time to throw in the towel. This dog won't hunt anymore.

There are lots of ways to get out of this hole, but all of them involve a change in business strategy. Again, I'll point you to some great reading. Read the third section of How To Save Jobs. It will show you how to evaluate your assets and reinvent your business from the inside out.

Live long and prosper

I sometimes get the urge to go out and build another software product. I love to code and I actually love the entire chain of activity that is product marketing. But after more than two decades of running my own software and then publishing company, I also needed to get a life.

I was fortunate in that I was able to sell off my software assets back in the day, and have managed to transform my career so that I get to talk to you here on ZDNet, lecture, advise, and teach. I'm having more fun in my work life than I've ever had before, I get to spend time with my wife, enjoy my car, lift weights, and sometimes work on my house and hobbies.

The point is that there is life after software. While my advice here is intended to keep you in business, if that time comes, you should know that there is life after software entrepreneurship.

Good luck. Live long. And prosper.

I grew up in the 1960s and 1970s. I never wanted to be a computer scientist or a programmer.

I remember that from about age 13 until I went to college, I had a big poster of the moonshot lunar insertion trajectory hanging over my bed. I liked science, and generally thought I wanted to be a scientist. As a kid who grew up in the heady days of the Apollo program, I had a vague idea that I wanted to work for NASA.

As it turns out, I now live on the Space Coast, about 40 minutes south of Cape Canaveral. Ironically, in the eight years I've lived here, I've never had time to visit the rocket site.

I was technically inclined, and so I spent a lot of time with the A/V club. I liked tinkering with the mechanical components of the projectors, and liked the sound and fuss of the sprockets as they wound film through the mechanism. I also liked avoiding some classes.

My public high school had a small mini-computer, a Digital Equipment PDP-8e. We also had some sort of tie-line with a teletype to a local college, but I spent all my time on the PDP-8. I don't recall if computer class was an elective or required for everyone, but I vividly recall the class itself — better than I recall much else from that far back.

One thing I recall was that I wasn't particularly excited about the computer back then. While the Apple II came out that year, we didn't have access to it. We had a mini-computer and it was anything but friendly. It had blinking lights, a front panel where you would toggle in boot code, and you saved your programs on paper tape.

One other thing I recall was my teacher. His name was Ron Mezzadri, although we called him "Mezz." I remember his method for explaining how computers were very literal beasts. He had some of us stand in the back of the room and told us to walk forward. When we avoided all the desks, he told us we weren't thinking like a computer.

Then he had one of us give him instructions. When he was in the back of the room and was told to walk forward, he bashed into the desk and stopped. It took us most of the class session to discover we needed to very carefully guide him, step-by-step, turn-by-turn, all the way across the room.

He taught us not just how to program in BASIC, but the basics of assembly language. Since the PDP-8 needed a boot loader to be toggled in to boot up, he taught us what that meant, what binary code was, and showed us how to quickly toggle in the boot loader to start the PDP-8.

Later, when I went off to college, I was in school near Maynard, Massachusetts, where Digital Equipment Corporation was located. They made the PDP-8. At some point during my time there, I somehow (that memory is lost to time) got my hands on a raw front panel for a PDP-8 (the holes where the lights and switches would go were never cut).

I'm bringing this up because my wife talked me into helping her clean the garage this weekend, and I found that front panel among the items stored on our shelves. I'm going to need to dust that thing off and hang it somewhere. It's pretty special.

In any case, back to programming in high school. My teacher told us about a weekend program at the New Jersey Institute of Technology. It was a first-year college programming course in Fortran, but it was being offered to select New Jersey high school students. I applied and was accepted.

The "gotcha," at age 15 or so, was I didn't have a car and Newark was a 45-minute car ride down the Garden State Parkway. My dad was a hero. He drove me down there, on Saturday mornings, and came back four hours later to pick me up. That man had to spend three hours in the car every Saturday for sixteen or eighteen weeks, just so I could take Fortran.

For whatever reason, I was a natural. I "grokked" programming. I just got it. But, to this day, I couldn't tell you if I just "got it" because of some twisted spark inside of me, or because of how Mezz taught the class and triggered something, a something that has since been a big part of my career for all these years.

I wound up leaving high school a year early. By the end of my junior year, I'd used up all the school's science and math courses, and having taken (and aced, thank you very much!) a for-credit college Fortran course already, it seemed to make sense to skip my senior year.

That took some serious 'splaining, the day I decided this was what I wanted, and then had to convince my parents. But they eventually gave in, I applied to engineering school, and I got in — the youngest student in the college.

With the fog of time, I couldn't tell you whether I went to college early because of my programming classes, but they certainly contributed to the idea. After all, having finished my first college class at age 15, it seemed like something I could do.

The odd thing is that once I was in engineering school, I tried my darndest to major in something other than computer science. I believed that computers were a tool you used with other sciences, but I didn't just want to use them alone.

I started as a nuclear engineering major, switched to mechanical engineering, and for my first two years, tried the basic engineering courses for almost all the majors. They did not come naturally. They were a struggle.

Meanwhile, by the time I hit my junior year in college, I was taking the graduate school courses in microprocessor design and computer graphics. I'd used up all of the undergraduate computer science curriculum and by senior year, I was actually co-teaching the graduate microprocessor design course.

Also, by junior year, I still wasn't a computer science major and I was still trying to figure out what I was going to study in college. Eventually, my academic advisor broke it to me as gently as he could: I might not have declared as "computer science," but since I'd completed the computer science curriculum (and pretty much nothing else), I was, by default, a computer science major, whether that was my plan or not.

This week, ZDNet's Charlie Osborne and Matt Baxter-Reynolds are debating Should kids be taught to program?

I can't tell you whether all kids should be taught to program rather than, say, as Matt put it, "how to read mass spectrograph output...or how to calculate stresses on a suspension bridge."

But I can almost definitively tell you this: if my public high school didn't have a teacher like Mezz and hadn't introduced me to programming while I was still a kid (and this was the seventies, remember), I probably would never have had the incredibly exciting and gratifying career in computer technology that I've been so fortunate to experience.

Oh, and I still code. I also teach programming. Using Mezz as a role model, I teach object-oriented programming at the University of California, Berkeley extension. One of my most fervent hopes is that I'll inspire one of my students (who are mostly adults looking to improve or change careers) the way my teacher inspired me all those long years ago.

So, should kids be taught to program? Given that computers are far more accessible than mass spectrographs or suspension bridges, and given that there's a much larger world of opportunities for making money today both inside and outside the corporate world (can you spell app store? Sure, I knew you could) I have to answer with a resounding, "Hell yeah!"

While I can't directly recommend the purchase of a Surface RT over an iPad or Chromebook, or a even Surface Pro over a similarly equipped ultrabook, there are some features in both Surface models that are better, more innovative, and potentially more game changing than on the iPad.

Way 1: Active tiles

Say what you will about the iPad's launcher interface, it's definitely long in the tooth. In fact, it's pretty much the same interface that Apple has been using for 20 years.

The nice thing about Microsoft's much-maligned Start screen is that the tiles dynamically update. That may be a pain for a desktop user, but on a tablet, it could be a big help.

On the one hand, it's almost impossible to tell which tile will launch YouTube, because the tile has a picture of whatever was recently watched or is currently being featured. But, on the other hand, dynamic tiles used in, say, a medical environment. could give professionals an at-a-glance view that the iPad launcher can't come close to duplicating.

Way 2: Named sets of applications in the launcher

The iPad lets you group sets of applications into folders, but the icons in those folders are impossible to discern to the naked (or at least, over 40) eye. The iPad also lets you stick a pile of apps on a page, but it doesn't give you the opportunity to name the page.

The Start screen allows you to group applications, and assign those groups labels, which appear above the groups. The icons remain full sized, but you can get a nice overview by simply zooming out. It's a very clean interface for a challenging problem, and it gets you active tiles, named groups, and a bird's eye view, all in one clean interface.

Way 3: Ability to run apps side by side

You can run two apps, side by side, on one screen. No more flipping back and forth like on the iPad. Granted, it's not real windows like, you know, Windows, but it's a huge improvement over the Fisher Price iPad interface.

Way 4: Real Microsoft Office

As our own Mary Jo Foley reported, it may even be a longer wait than we all thought for a real version of Microsoft Office to arrive on the iPad.

There's no doubt that Microsoft's licensing is twisted and sick, as evidenced by the fact that Microsoft includes a product called "Office" on the Surface, but you're not allowed to use it in, you know, an office unless you buy another license. It boggles the mind.

Even so, if you want real, true Microsoft Office, you can't get it on an iPad. You can on a Surface device. One ding against Microsoft: There's no Outlook on the Surface RT. Seriously? What could it possibly have been thinking?

But if you want the real-thing Word, Excel, and PowerPoint on your tablet, you'll want a Surface, not an iPad.

Way 5: Actual USB

Both Surface variations allow you to plug in some (not all, but at least some) USB devices. That means you can finally put files on a thumb drive and move them onto your tablet. You can also use an external keyboard and mouse, so you no longer have to smush your fingers all over the screen to get anything done.

You'd think Apple would have figured this out years ago, but instead, it effectively ceded its entire file system to Dropbox.

Conclusion

So there you go. I can't tell you to go out and buy a Surface (especially the Surface RT), but there are some good ideas in the design that the iPad is sorely lacking.

My hope is that Apple will innovate and move its launcher interface out of the 20th century, and that Microsoft will do a better job of both meeting tablet owners' needs along with the needs of the rest of its traditional Windows users.

Yeah. Well. I know.

Chrome just recently decided to change its menu behavior to accommodate fat fingers.

Menu items are now separated by extra white space. For those of you with touchscreens, that's probably a big help. But for those of us using a mouse and a keyboard, all this update does is use more screen real estate.

For those of us with big lists of bookmarks, this can be really annoying. The following image shows the difference before the update and after.

Fortunately, there's a pretty simple fix, thanks to a post on the Google Chrome forums.

Simply open up Chrome's properties, and add the string "--disable-new-menu-style" as shown below:

I have to agree with poster CorteXodus that not having an obvious way to turn this off is ridiculous.

Some of us still need to use real computers.

This fix works in Windows 7 and Windows 8. I'm not sure about the other operating systems, but at least it's a start.

If you're using something else and you can use a modified version of this solution, or you find another, post it in the comments below.

Update: As reader Michael Alan Goff suggested below, there are about:flags options available called "Enable touch events" and "Touch Optimized UI". On my Chrome, these are set to automatic, but setting them to Disabled may also provide the solution. This is just one of the reasons our readers are so very cool.

Update 2: Well, that didn't work. I tested it on my Win7x64 box and the about:flags hack didn't work, but the command-line parameter did. So, that's your path for now.

Update 3: My menus went back to fat-finger menus today. However, when I closed Chrome, right-clicked on the Chrome icon in Windows (both 7 and 8), selected Properties, closed the dialog, and then re-opened Chrome, the normal menus came back. I'm really hoping this isn't going to be an ongoing nightmare. Otherwise, maybe it's time to reconsider Firefox or (wow) IE.

Today's iPhone 5

There's been a lot of discussion recently about Apple's innovation gap as compared to the other smartphone and tablet makers. Innovation, the story goes, has slowed.

There is no doubt that mobile devices have evolved considerably — especially in terms of hardware capability — over the past few years. The decline of the traditional PC market is a testimony to that fact.

Even so, there's one area — and an important one at that — where Apple barely innovated in almost two decades: The main launcher screen for mobile devices, now found in its iPhone and iPad products.

I know what you're thinking. The iPhone is only six years old. Granted, the little icons on the screen haven't changed much since the original iPhone was launched (for comparison, Microsoft was shipping Windows Vista back them), but, even so, six years isn't 20 years.

Maybe so, but climb onboard for a short tour back in time. When we're done, I think you'll agree that one of the reasons everyone is so incredibly bored with the iPhone and iPad launcher interface is that it's all been done before.

Climb into the DeLorean, get your speed up to 88 miles per hour, and we'll work our way back to a time when Bill Clinton had just become president, Michael Jackson was still in his prime, putting on a record-setting show at Super Bowl XXVII, Jurassic Park and Mrs Doubtfire were tops in the movie charts, Microsoft was selling Windows for Workgroups 3.1, and Mark Zuckerberg was 11.

Today's iPad

This is today's iPad. Notice two main characteristics: The set of four main icons on the bottom of the screen, and the grid of app icons in the main screen.

The iPad, of course, can have up to six icons in the main app section. Remember this when we get back to 1993.

Today's iPad mini

For completion's sake, let's also look at today's iPad mini — which is identical to today's iPad. On the iPad mini, you can also have up to six main application icons.

2007: The original iPhone

It's 2007. George W Bush is president, the housing bubble hasn't collapsed yet, Lehman Brothers is still flying high, and Apple introduces the exciting new iPhone, complete with an interface that has four main application icons, and a grid for secondary icons.

1996: Pilot 1000

It was 1996. Spin City and Sabrina: The Teenage Witch were debuting on TV. It was a great year for bad blockbuster movies, with Independence Day, Twister, and The Rock dominating movie screens the world over. And US Robotics (not yet even called Palm) had introduced the Pilot 1000 and Pilot 5000 (the name "PalmPilot" was a trademark lawsuit still in the future).

While slightly different in form from the iOS devices, the Pilot had four main application buttons along the bottom of the device (hard buttons, this time) and a grid of application buttons on the main screen. The Pilot, famously, also used a weird form of handwriting recognition we all got to know for about a decade called Graffiti.

1993: Newton MessagePad

As the DeLorean comes in for a landing, we discover the original Apple handheld device, the original Newton MessagePad. The MessagePad actually shows six main icons (hmmm ... our current iPads will also support six main icons) and a grid of supporting applications.

It seems that the more things change, the more they stay the same.

(Image: ZDNet)

Present day: Windows tablet

Finally, just as an exercise in contrast, I present to you the Windows 8 tablet. Doesn't look at all like the old Pilot, original iPhone, or 1993 Newton, does it?

I guess it is possible to innovate a little when it comes to interface design. Granted, the Windows 8 device has its limits, but there's no doubt that Apple's launcher interface is long in the tooth.

Oh, and no, this isn't about Linux. Surprised?

Over the past two weeks I have continually lowered my requirement set, reduced my "must haves," given up on my "like to haves," to the point where there's nothing left, not really.

It's about image management. Sigh. Don't start telling me image management is easy, Picasa, yada yada, Elements Organizer, yada yada, Lightroom, yada yada, Bridge... whatever. I've heard it all. I've tried it all.

I give a lot of presentations. A very lot. I spend hours, days, weeks, months of my life in PowerPoint. No need to pity me. I actually quite like PowerPoint. But the point is, to make these presentations more interesting and explain things more clearly, I use a lot of images.

Not just photos. Images. And here's where things begin to break down.

There are two completely different classes of images out there: bitmap-based images and vector-based images. Photos are bitmaps, filled with lots and lots of pixels of information. The more pixels you have, the higher-resolution the image.

Vector-based images are line drawings with fills. Rather than huge matrices filled with dots, vectors are actual line and curve formulas, linked together in a format that describes an illustration.

Bitmaps only scale if you have a boatload of bits. Vector images scale naturally, because the formula just recalculates for the larger size. Vectors, therefore, are ideal for drawings and illustrations, rather than photos and paintings.

Photoshop does bitmaps. Illustrator does vectors. Photo file formats are things like JPEG, TIFF, PNG, GIF, RAW, and so on. Vector file formats are things like EPS and AI.

With me so far?

I have thousands of images, both bitmap and vector. In the case of photos, I've taken quite a few myself. I've also bought a lot of stock images. In the case of vectors, I've bought most of them, but modified some of them in Illustrator and Photoshop to best suit my presentations.

The problem is, finding the right image has been getting out of control. It can take an hour or more to just find an image in my library. Yes, I've organized my folders as best as possible, and I can easily dig through and see thumbnails, but I wanted a better way.

I wanted to search by keyword, review images quickly, search across collections, choose based on metadata, file type, and more. I wanted a way to find an image in 30 seconds instead of 30 minutes. I wanted an image organizer.

Here then, are the simple set of specs I started out with:

• I wanted to have a database-based organizer, so that searches would be fast and all the files wouldn't have to be scanned for each search.
• I wanted that database to hold all my media asset files (both vector and bitmap).
• And I wanted that system to allow relatively easy drag-and-drop from the desktop to the application so I could get content in and out of the system while composing presentations, without losing track of the flow of the actual lesson I was preparing.

Oh, and it would be nice to have this on a network, so I could easily do my work either at my desk or on my laptop.

I am in a foul mood. Enter the photo organizer category.

There is a very large category of software called the photo organizer. As you might imagine, these products organize photos. Right away, you can see the problem, right? Photos. Bitmaps. They know nothing of illustrations and vector graphic files.

I set these products aside for a while as I searched for a more comprehensive asset management tool. In terms of standalone products, the only one I found was called Portfolio from Extensis. This product hasn't been updated since 2011, and its main version number hasn't changed since something like 2005. The company also didn't respond to requests for information.

There is a category called "Digital Asset Management" out there as well. These are enterprise-level products, often Web-based. You can begin to tell they'll be trouble because there's no price for the product on the site. Almost all providers of DAM tools have a "let us have an expert call you" button.

Here's a hint: I have two remaining days to implement the entire solution, and if I have to have an expert call me (apparently, a unique phrase for DAM solutions), then I can pretty much be assured it's not going to save me time (and most likely will be way outside my budget).

I even tried SharePoint. I have Office 365, which comes with a SharePoint account, and SharePoint has a Media Asset Management application. Feh. It will store files, but it is about as interactive and drag-and-droppy as a dead fish. On top of that, PDF files loaded into SharePoint's media management system show as default files. They couldn't even be bothered to render a thumbnail of the PDF.

Plus, after two or three calls to Microsoft, no one could answer whether or not it's possible to grow my 10GB SharePoint disk quota to a larger capacity, if I needed it to store more images.

Major fail.

So I decided to lower my expectations. Rather than organizing all my images together, I'd create one bucket for bitmaps and one bucket for vectors. After all, the market is clogged with photo organizing tools.

The Big Kahuna of photo organizing tools is Adobe Lightroom. I'll get back to that in a minute. First, let me talk about all the other photo organizing tools, with a particular nod of "WTF" to Google's Picasa.

Most photo organizing tools try to be editors as well. I just ignored that part. I didn't want red-eye correction, I wanted to find pictures with eyes in them. But most of the tools will catalog JPEGs and camera RAW. Some can handle PNGs and TIFFs, and a few handle old-school GIF images.

With one or two exceptions, the database catalog in these systems resides on your local system. This is a single-user application, and if you try to move the database somewhere else, there's no end of complaining on the part of the application.

Next up, I rant some more, talk about Picasa's weirdnesses and Lightroom's fail, and rant even more. Good times, good times...

Picasa is special though. Picasa doesn't just open once you install it. Picasa gives you a choice: scan your entire hard drive, or just major folders. There's no "don't scan because I want to tell your where my stuff is" option. So Picasa scans. It found hundreds of folders filled with various parts from previous projects, none of which I wanted in my database.

What gave me the willies about Picasa, though, was its insistence that it wanted to share my images on Google+. I avoided this by refusing to log into Google with Picasa, but it made me nervous. I have a lot of licensed images I'm allowed to use in my own works, but not that I'm allowed to publish as standalone images. I didn't want Picasa to just upload all my files, willy-nilly.

And, besides, Picasa only handled bitmaps anyway.

With most of the other photo organizers, there's really not much of a difference. They all were a bit sucky in one way or another (I downloaded and tried at least eight of them beyond the usual suspects), and since there was a well-regarded professional option, I decided to discard the consumer products and look, once again, to Adobe.

This is where Lightroom comes into the picture.

No matter who you talk to, when you start talking about professional photo management, Lightroom comes up. So I bought Lightroom. I found a good one-day discount deal, but still, there's a hundred bucks down the toilet.

Lightroom is a little more fussy than some of the other photo organizers, in that it doesn't recognize PNG files. PNG files have pretty much replaced GIF images in Web design, are used all over the Web, and have the happy property of being able to be transparent, so if you want to stick a person in a presentation, you don't have to bring the background along with it.

Lightroom doesn't like PNG. Apparently, PNG doesn't handle metadata in a way that Lightroom considers robust enough, so if you have PNG images, well, you're just out of luck.

I don't need PNG images, but I do need transparent images. TIFF also supports transparent images and Lightroom supports TIFF. So I bought a couple of batch converter programs that convert from PNG to TIFF. For those of you keeping track, there went another hundred bucks, flushed away.

Can you see where this is going? Lightroom couldn't read the TIFFs produced by the conversion programs, although Photoshop could. But if I couldn't organize them in Lightroom, then I pretty much couldn't use Lightroom. And yes, I could convert the PNG images to transparent TIFFs in Photoshop, but even with some of Photoshop's batch settings, my desire to save time managing images was rapidly becoming a second full-time job (well, technically, it would be a fourth full-time job, but who's counting?).

And I haven't even talked about the curation process required to assign appropriate keywords, and so forth. We're not even there yet. That's another nightmare and lifetime of organizing, all on its own.

If you've been following along, you've figured out that Adobe's gift to professional photographers, Lightroom, couldn't cut the mustard with my project. Not only couldn't it handle vectors, it also couldn't handle PNGs or converted TIFFs.

Exit Lightroom, enter Bridge.

Bridge is the other "must-have" Adobe solution for managing files. The win with Bridge is that it can read all the different file formats, including the vector formats. Yay that.

But Bridge doesn't have a central database. Every time you enter a directory and want to see the images in it (or in its subdirectories), or search the images in it, Bridge does a new scan. I selected one collection of a mere 2500 photos and I tested Bridge on it. Ten minutes to do a scan. Leave the directory to look somewhere else and come back. Ten minutes again.

This, essentially, makes Bridge too slow to use. But the kicker was that Bridge won't show transparent PNGs as transparent.

It insists on putting a white background on them, so it's impossible, visually, to tell which PNGs are transparent, and which have a white background, unless you go ahead and open the image up in Photoshop -- which defeats the whole purpose of trying to use an organizer tool for quick image searches.

So where does this all leave me? Cranky. Very, very, very cranky.

Lightroom can't organize vectors, won't read PNGs, and won't understand converted transparent TIFFs. Bridge takes a lifetime to scan large directories of images (which is the whole point) and won't display transparent PNGs as transparent. The other photo organizers break in similar ways, and won't display vectors.

Worse, the curation process — trying to sort out, label, and keyword thousands upon thousands of images — is a job for a team of people over six months, not one lone professor working in the comfort of his home office.

I have to say that it astonishes me that there isn't a widely available solution to this. I'm flabbergasted (love that word!) that Adobe doesn't have just the perfect solution to this problem (beyond the somewhat anemic Bridge) because solving this sort of image management problem is at the heart of what Adobe does.

I'm also disappointed in the Web-based and enterprise-based solutions.

First, the barrier of entry is huge. There appears to be a disconnect between the needs of a professional designer with thousands of images and a large corporation buying an enterprise package.

Second, most of the Web gallery and enterprise solutions still use relatively primative upload dialogs and download buttons. There are very few solutions that will let you drag from a Web page into a desktop application, or to the desktop, and do it for a bunch of images, and those that do also seem to think the only type of image that exists is based on bitmaps.

Say what you will about cloud computing, the true, native desktop application is still better at some types of interaction.

So, I've given up on this project.

I have had two days until I need to start my next major series of PowerPoints, and I'll be doing them straight for about six weeks, so there's clearly no way I can solve this in two days. I'll probably spend a day doing a little disk file system housekeeping, but other than that, my search for media asset management is a complete bust.

Did I mention that I'm cranky? I want a cookie.

UPDATE: Well, I haven't completely given up. I never really do when there's a problem to be solved. I'm working on a new approach. Rather than look for a good media asset management tool, I've decided to focus on super-charged file managers, and see which ones are available with good media extensions. I've been looking at one over the weekend and it shows promise. So, stay tuned. There's probably another article on this topic coming real soon now.

I still want a cookie.

But, until now, I haven't explained why I chose to migrate to Office 365 and not Google Apps. I'll tell you this: it wasn't an easy choice, and for some surprising reasons.

Some background

I've been using Exchange since 2002, when I started the OutlookPower Web site. Prior to that, we were actually using the incredibly easy-to-administer Mac-based Eudora mail server, which ran on a Mac. But when ZATZ decided to add an Outlook-centric site, I thought it would be a good idea to eat our own dogfood. To become familiar with Outlook and Exchange, we moved from Eudora to Outlook.

By the way, this wasn't the first time I did the dogfood thing with our email environment. Way back in the mid-1990s, I was the editor of the print newsletter Insider for Lotus cc:Mail, produced by the Cobb Group, an ancestor company of ZDNet sister site TechRepublic.

Way back then, since I was editing a publication about cc:Mail, I decided to use it for the company mail server. That was a wacky experience. The cc:Mail environment ran on DOS (yes, I'm not kidding), and each piece of the mail server (the database, the incoming mail exchanger, the outgoing mail exchanger) needed to run on its own physical server box. It took four or five physical machines to run this one mail server. Talk about technology in need of virtual machines, eh?

So, jump forward to 2002 when I was now runing a site about Outlook. I decided I'd use Exchange. Exchange is a lot like chess: relatively simple to learn, but takes a lifetime to master. I didn't quite realize that at the time, so I set up an Exchange server, moved us all to Outlook, and was happy ... until the day it all went to crap.

My Exchange server blew up. Literally, the drive inside the server exploded into dust (literally).

 Unfortunately, while I had lots of backups, I didn't know (or at least take seriously) the warnings about needing a second Exchange server on the network for Active Directory continuity, and when my sole Exchange server (and domain server) turned to dust, recovering it was a 13-day nightmare.

Long story long, after that, I decided I'd rather let an expert manage my Exchange server. An IT friend of mine recommended a regional hosting provider that he'd worked with for years and found reliable. They also had two additional benefits: $9.95 per month per users, and unlimited mail storage.

So I switched. Since then, I've been using this regional hosting provider for 11 years, quite happily.

Unfortunately, in the past six months or so, the company changed its business model (probably in response to Office 365), de-emphasizing Exchange and adding other hosting services.

My Exchange email flow reliability has begun to deteriorate.

Email is my income's killer app. I need to check it constantly to be responsive to projects I'm working on. Three hours of downtime is deadly. I could slip on something important or not respond to an urgent request. My hosting provider started to have mysterious outages. Then, their support site changed, and all the materials on the support site were unrelated to the sort of account I had. You get the idea.

It was time to look for another provider.

The case for Google

The case for Google was pretty simple. The two organizations I work most closely with, CBS Interactive and UC Berkeley, are both Google shops. I am in contact with my CBSi colleagues constantly — all day, every day — and I have some interaction or another with the University of California each day as well. The nonprofit I work with is also a Google grantee.

As part of my daily work, I often need to connect into CBSi or UC Berkeley files on Google Docs. I get sent links that require a Gmail account ID, and so forth. The case for Google is the theory was that if I moved fully into the Google world, then all this would become much more smooth.

It would also avoid the occasional problems of someone sending me a message directly to my never-checked Gmail account, instead of my main email account, since they'd be one and the same.

Another plus is that I rely constantly on Google Calendar. I've managed to make that application jump through some amazing hoops and have it customized in such a way that it's one of my most critical management applications. I use it not just for appointments, but for project management as well.

Google would be slightly less expensive, too. At $50/year per seat, it would cost us $100/year for my wife and me, vs. the $240/year we're spending now, or the $360/year that Office 365 would cost.

Some of that savings would be lost, because I use PowerPoint constantly for work, and I'd have to buy PowerPoint licenses no matter what. I'd probably also need Word, since I do a lot with collaborative review markup in Word. While we have a bunch of Office 2010 licenses, I'd probably need to buy at least two new Office licenses, at least sometime in the next few years.

But still, the idea of going all-Google, all-the-time was attractive.

Next, let me tell you the case for Office 365, and then I'll tell you about the big deal-breaker with Google that made the Google choice a really bad idea.

The case for Office 365

Let's start with the big one: we've been using Outlook for 11 years, we know it intimately, we're comfortable with it, it works, and all our data, rules, customizations, and more are Outlook-centric.

Some of you may not like Outlook, but I actually like using it more than Gmail. But, as a test, my wife set up a Gmail account for one of her hobbies, and used it to correspond with other hobbyist friends, and discovered that she just didn't like Gmail much, either.

So, from the personal preference point of view, Outlook still had a lead.

Then there's the whole Office thing. My wife and I each regularly use three copies of Office on the three machines each of us use (that's a total of six licenses). Although our current Office license would do for a while, I really wanted the latest PowerPoint (the merge-shapes feature was the big draw).

When you sign up for Office 365, you can get Web access to the Microsoft apps for a lower price, or five licenses per user for the real, desktop Office apps for $15/mo per user. Since this was only $5/month more than we were now paying just for email — and it also added SharePoint, Lync, and a variety of other services I'm not using now but probably will — it's actuallhy a pretty good deal.

Moving to Office 365 would be relatively straightforward since our existing Exchange environment could be migrated to the new one and all our accounts would pretty much stay the same.

Once moved to Office 365, life would remain essentially unchanged compared to the way we worked before migrating.

The cost case

There wasn't a huge cost difference. Office 365, for the plan I chose, would cost $120 more per year over what we spend now, for both of us together.

If we went to Google Apps, and then bought a home Office license, we'd save $40 a year over what we pay now. But since I use Office for business, I'd be uncomfortable with a home license to Office, so we'd wind up paying more — pretty much around the same as with the Office 365 plan.

The net-net was that the cost difference between the two plans was ultimately so negligible, it didn't factor into my decision at all.

The big factor, the really big factor, was the Google account problem. That's next...

The Google account fiasco

Explaining this is going to make my head hurt. Everything about Google accounts makes my head hurt. Let me preface this by saying I'm probably going to get some of the details wrong. Google accounts are so convoluted that the details are quite confusing.

If you're thinking about going here, you'll need to do your own research, not just rely on what I'm telling you.

Here's the key gotcha: if you move to Google Apps, you can't take your individual Google account into Apps.

Explaining this is going to make my head hurt. Everything about Google accounts makes my head hurt

That means that whatever Gmail email account you may already have, the one tied to your calendar, chat, incoming Gmail, your YouTube favorites, and on and on and on would not be tied to whatever account you get for Google Apps.

It's possible to do some finagling.

For example, you could share your existing Google calendar with your new account and give it read/write access. You could set up a forwarding filter from the old Gmail account to the new one. You could grant access to your Adwords and Adsense settings to the new account.

But other things aren't so straightforward. Have a ton of contacts in Google Chat? You'd have to go and re-invite each one to your new account. You couldn't use your existing Gmail account name, so if you happen to like your existing Gmail account name or if it's something people are long familiar with, say goodbye.

And if your existing Google account is a member of a Google Group or has sharing privileges on Google Docs documents, all of those would have to be changed to your new Google account. Add to that phone numbers attached to Google Voice, and you have a real mess on your hands.

Yes, sure, you could still use your old account, but if that's the case, and you're trying to go for an integrated, holistic whole, having to constantly switch between a new Google Apps account and a legacy individual Google account shoots the desire for a well-oiled, cohesive messaging infrastructure right out of the water.

Throughout the whole Google Apps environment, there are little gotchas about how you transition your account, and what you can and can't do when moving to an Apps account from an individual account.

Personally, I think this is ridiculous.

Most people will likely have started with individual Google accounts and gotten to like Google services, and then decide they want to move to Apps. To force people to either lose all that infrastructure, or have to do a whole pile of hoop-jumping to make it work seems wildly counter-productive.

How I made my choice

In my case, I realized it would be far simpler to move my Outlook from one Exchange hosting provider to another.

If I instead moved to Google Apps, all my long-standing Google Chat connections would have to be re-invited. I have a ton of these and I didn't relish having to explain this whole process in gory detail to the half of them that would demand some sort of backstory explanation. All of the projects and groups I'm involved in with for work would have to be changed, which would mean I'd have to assign twenty or more administrivia to-do items to some already extremely busy people.

Plus, there was the chance everything could go wrong. I had a situation a few years ago where I had a Google account, gave it a backup email address, and it decided to make some severe changes to the original Google account because it already had the backup email account listed with some other service. That situation was never fixed, because the Googlers I eventually managed to talk to said there was nothing that could be done. It was just what happened.

Add to that the fact that neither my wife and I particularly like using the Gmail interface (and given I spend so much time in my mail environment, that's important), and that there was really no cost advantage when factoring my specific need to purchase Microsoft's Office anyway, and the decision was becoming clear.

What really clinched the deal for Microsoft and against Google was the homework problem. I didn't want to have to be a squeaky wheel to my busy CBS Interactive and UC Berkeley colleagues. I didn't want to make them all change my chat names, change my account access to all our shared documents, and make them re-invite me to all the groups I rely on to do my daily work.

The bizarre moral of this story is I would actually have less service interruption and more service continuity for my existing mission-critical Google services if I switched to Microsoft Office 365 than if I switched to Google Apps.

And that's why I switched to Office 365. So far, two weeks in, it's all running quite smoothly, both my Google accounts and my email on Office 365.

I have one more story coming in this series: how I moved a half million email messages between services. That's coming next week.