Tom Petrocelli's take on technology. Tom is the author of the book "Data Protection and Information Lifecycle Management" and a natural technology curmudgeon. This blog represents only my own views and not those of my employer, Enterprise Strategy Group. Frankly, mine are more amusing.

Friday, August 22, 2008

Ditch the Computer for the Net - Not!


I have been fooling around with Internet desktops for a little while now. Two in particular have floated to the top of the heap, Desktoptwo and Jooce (pronounced like juice). While they have the same fundamental goals, the approach is quite different.

When looking at Internet desktops it's important to understand why anyone would want these. Basically, they claim to allow you to become free of your desktop computer or laptop. In other words, you can access your "Desktop" from any browser, presumably even on a smart phone. That is an attractive option for people who need to access files and applications from different computers and devices. If I'm visiting my family in Florida, I could, in theory, access my desktop remotely. But, as Yogi Berra once said, "In theory there is no difference between theory and practice. In practice there is." So, I'm skeptical.

Both Jooce and Desktoptwo (and all other Internet desktops) allow you to have personal communications and storage in your virtual desktop. That seems to be the core of it. It sounds like a nice idea until you consider the following:

  1. Lot's of options exist for accessing files through on-line storage applications. Dropbox and XDrive are two I use regularly. Specialty services also exist such as Flicr and Google Picasa for pictures. Do I need another place to store and share files?
  2. They continue to rely in some fashion on local machines. Jooce especially only has the ability to display or run media files such as audio, video, or pictures, and do instant messaging. Desktoptwo comes with more applications including a version of OpenOffice and a PDF reader but no video capability. Applications can't be installed on your on-line desktop. You are completely dependent on what is provided as part of the desktop or some other site on the internet. If I can keep my documents on Google Docs or Zoho, and my pictures on Flicr, do I need this?
  3. The same problems exist as with all on-line applications, especially its unavailability if you are lacking a network connection or the desktop service company's servers are down.
  4. I don't see how you can make real money. Desktoptwo has ads that are exceptionally annoying but I never read them. The fact is, they seem out of place on a desktop. This was true ten years ago in the age of Pointcast and other push media. Jooce doesn't have any ads at all. Maybe they hope to sell the software to ISVs...

There are real differences between Jooce and Desktoptwo that, unfortunately, highlight the problems of virtual desktops. To that end, I welcome you to enjoy these mini-reviews of each. Keep in mind, both are officially in Beta so, to be fair, these are preliminary comments.

Jooce

I like the look and feel of Jooce. It has a Gnome-like (as in Linux/Gnome interface) look that, while a bit cute, is at least visually appealing. You can customize the desktop a bit with your own wallpaper which is nice. Jooce actually gives you two desktops - a private one and a public one. The public one allows you to share files with other Jooce users. Switching and sharing files between them is easy.

Jooce is very media-centric. While you can upload (there's that tether to the local machine again) and store any files you want it only has applications for audio, video, and graphics files. While this is very limited, they are integrated into the desktop seamlessly. For example, if you open a picture, it appears on you desktop as a photo or picture that you tossed on your desktop. Neat! Audio just plays without appearing to launch a separate application. This is the true innovation of Jooce. It strives to create a desktop where applications are part of the desktop and not separate from it. This is also a weakness since the number of applications is limited (to the above add an IM application that can connect to major services and that's all) and even if available would be hard to integrate so seamlessly. They are promising e-mail soon but even viewing a PDF requires that the viewer be part of your local browser environment and hence your local machine. Go to a machine that doesn't understand PDF and where you are not allowed to install a viewer and you can't access your files.

Jooce has promise as a way to store and share files but it is still anchored to the resources available on the local machine a bit too much.

Desktoptwo

Desktop two has a more conventional look and feel. It has lots of conventional looking (kind of a KDE-ish) icons on a conventional looking desktop. It is best described as flat and uninteresting. It does have more applications including the OpenOffice.org suite, e-mail, calendar, RSS reader, IM, a website editor, blogging tool, and a PDF viewer, though I can't find a video player.

It is unbelievably slow. Slow to start up, slow to load applications, and slow to upload files. Worse yet, it often appears to be uploading files but they don't appear on the "hard drive". It appears that it it can take so long to transfer a file, especially a large one, that it is unclear that it is actually happening. This makes it too easy to start an upload, assume that something mysterious has occurred and shut down the desktop before it can fully upload the file. And the ads are really annoying.

The interface is also Flash-based which generates weird error messages. For example, while trying to empty my trash bin I got an error message telling me that the script was taking too long in this "movie". I know that is because of the roots of Flash as a media scripting applications but it is out of place here.

So, Jooce hasn't enough juice yet to be useful and Desktoptwo, in trying to be a full fledged computer in a virtual, on-line space, is slow and ponderous. The edge goes to Jooce, however, for its look and feel. In fact, I would like to have the interface on my regular desktop or phone. It is, at best, an adjunct for a real desktop but that's enough. Desktoptwo seems to be shooting for the corporate market for virtual desktops. It's got a ways to go then.

I like the idea of a desktop that runs in a browser which gives me basic functions no matter what platform I'm running it on (including a phone). On-line desktops have a while to go before that is a reality. It will be interesting to see how this plays out.

Friday, August 08, 2008

I Am Dumb

If you haven't heard about this already, then let me be the first to give you a chuckle. Apparently, someone uploaded an application to the new Apple iPhone store called "I Am Rich". Priced at 999.99 (the highest price the store allows) it does basically nothing. If you click on its icon, which is shaped like a ruby, it displays a picture of a bigger ruby. The author makes no attempt to hide the fact that this application doesn't do anything. He is right upfront with this being $1000 for something completely useless. The application's sole reason for being is to display to the world that you are rich enough to toss down $1K on nothing. It is the 21st century version of lighting your cigar with $100 bills. Channel Web had a great article about it.


Two things really caught my interest here. First, that eight people actually paid for this. That's incredible. Even if you are Bill Gates-wealthy you don't throw your money away like that. The more important fact to emerge in this whole farce is that some of these people were not rich and did it because they thought it was a joke. Are you kidding me? At what point after they entered their credit card numbers in did this cease to be funny. Hello! It's not a joke when you pay serious money for it.

This leads me to the real purpose of this rant - idiot proofing. There has always been tons of talk in the engineering profession about making something idiot proof. As this situation shows, it simply can't be done. By definition, smart people don't understand how idiots work. They simply cannot conceive of the stupid thinks some people will do. Internally, they dismiss the really dumb behaviors of some people as being impossible. To themselves they say "There's no way anyone would do that". They might not even be able to think of something to mentally edit out because it is not how they would act or think. Just look around YouTube and you can see all the stupid things people do that a reasonable person wouldn't even be able to wrap their head around let alone come up with on their own.

Simply put: You can't predict what an idiot will do. You can only react to what they have already done.

You can, however, predict what a reasonable person will do. Even if the person is inexperienced you know they will act in a reasonable manner. As a reasonable person you can understand and predict that. You can say "Oh. I can see how someone would do that."

So, I want to advocate a different design philosophy called "reasonable person proof". Strive to make products such that you keep a reasonable person from doing reasonable things that are somehow harmful. If this product is in an area you don't know very well, ask people in that area what they might do. Observe normal people using your prototype in normal ways. If you design hammers, watch a carpenter hammer nails into a wall and come up with a way to keep the shaft from shattering. Don't try to keep a moron from getting hurt when he uses the hammer on his big toe to see what it feels like. (If you don't believe me on that one check out this video of a bunch of morons doing just that. All I had to do was go to YouTube and type in "hit toe with hammer." For the record I only looked for the video after writing the sentence but I digress.)

Rather then try and keep all manner of stupidhead from doing something completely dumb and harmful with your product, recognize that you can't predict that. Put on all those warning labels that cover you legally but don't try and design around it. It's a fools game (pun intended). Instead, assume that your customers are reasonable people and design for them. They deserve that attention. The idiots do not.

Oh, and everyone email Apple and tell them to put "I Am Rich" back on the Apple Store. Whether it's a demonstration of fabulous wealth or a honey pot for dummies, they should let us make the choice including the dummies.

Wednesday, August 06, 2008

The Name Game


The great thing about kids is that they don't know all the things that that adults know. This makes them open to anything, any thought, any idea. They are not constrained by "the obvious".

So, I was taken aback by a question from my daughter the other day. She asked me "how do cars get their names?" I, of course, launched into a Marketing 101 lesson about names and brands. I started by telling her about how names are sometimes meant to describe and invoke an image. Others, I said, are meant to invoke an image or create an emotional response. This got me to thinking more deeply about names in the technology sector. Since technical marketing is a big part of what I do, this is not simply academic.

I mused on the names I've known. Windows, Macintosh, and the Linksys BEFW11S4 Wireless-B Cable/DSL Router.

Over the years I'm heard a bunch of great names and a lot of silly ones (I still think that del.icio.us is a stupid name). On reflection I began to realize that computer product names fall into a series of buckets.

So I present to you "Tom's Guide to Technical Product Names" complete with commentary.

The Functional

One of the most popular ways to name a technical product is by assigning a totally functional name. These names usually sound like a part number, such as XLP 3010. The purpose is to give a sense of industrial strength and encode some kind of information in it. With a naming scheme like this, you can easily tell that the 3012 has two ports and the 3014 has four.

Of course, you would never remember what the two or four ports are for. It is a conceit to assume that a customer will know your peculiar nomenclature. If you don't know what an XLP is, then the distinction between 3010, 3012, 02 3014 is lost on you. Just look at the IBM Storage products to get a sense of this. What does DS6000 mean? Nothing unless you know what it is already. And in case you think I'm picking on IBM, go look at Linksys product names. Some of these are meant for the SOHO market! What are they thinking?

By the way, would someone please tell LG that I can never remember that my phone is an LG-VX8350. And yes, I had to look it up. Razor is a much better name.

A subspecies of The Functional name is The Descriptive name. This type of name tells you exactly what the product is. For example, the HP BladeSystem is a blade server system. No kidding. The problem with The Descriptive name is that it's boring. You might know what it is but will you care?

The Image

Some products are all about image. They have fancy names that tell you nothing but sure sound cool. I will put Microsoft (a great name) Windows Vista in this category. This was one that my kids asked me about. When I tried to explain it was supposed to invoke an image of a horizon without limits they looked at me like I was ready for the old folks home.

The problem with image names is that they either connect or they don't. There's not a lot of middle ground. Over all, image names are the best. They are not boring and they catch your interest. That's why I like Jott. It gives you that sense of quickly jotting down notes and sending them to someone else. Memo would not have cut it. Descriptive but dull. Jott is cool without being obscure or cute.

The Consumer

Obviously for more consumer oriented products, more consumer oriented names are in order. Unfortunately, as an industry we don't do that well. Some work well like iPhone because they are part of an overall branding strategy. The reason iPhone works is because Apple spent years convincing us that "i" anything meant an innovative and cool-as-all-get-out product.

The drawback is when the name is too cute for technical people or grown ups to buy and not feel stupid about. This is probably why I won't sign up for Twitter. It's not that I don't get Twitter. I just hate the name.

That goes for color. Recently I bought a portable hard drive for someone in the company. The one requirement was that it not be in a color that embarrassed them. Finding one was harder than it should have been. Same goes for names. As much as I like the idea of the ASUS Eee. I can't imagine whipping it out in front of a customer and say "How do you like my new Eee?" Not happening...

The "Huh?"

Some names are simply incomprehensible. They make so little sense that you can only go "What were they thinking?" One of my favorite online services is like that, called "I Want Sandy". Sandy is supposed to be your online e-mail assistant. Get it? Sandy is your assistant so you want her there to help you. My apologies to the folks at Values of N but it's a great product with a name that sounds like a porn film. I'm surprised that more firewalls don't block iwantsandy.com.

Or the Dell Vostro laptop. Is that Serbian or did they just make it up? Maybe it's a Transformer. Even Inspirion was better since it invoked an image of being inspired in some way. How about the Archos media play. What's an Archos? A Greek Hero like Achilles? We know what happened to him. An old company I worked for used to pick names from foreign languages that invoked certain images. Since many people didn't speak those languages, the imagery was lost on them and the names mostly sounded silly.

I admit that coming up a product, brand, or company name is not an easy thing. Big companies spend millions trying to tease out a good name from fickle consumer attitudes. Still, the techie world seems to be especially bad at it. Perhaps it's the geeky nature of what we do. Wrong side of the brain and all.

It seems to me that we have to get more folks in the industry whose other side of the brain works better.