I think digital identity, or how we identify who we are are on the Internet, is the next HUGE obstacle, and opportunity, in the digital age. Doc Searls, who is always worth reading, points to a great talk by Dick Hardt on Identity 2.0, his name for the next generation of identity on the Internet. It’s a great, 15–minute introduction to a subject that will affect everyone who uses the Internet.
The news last week was all about Katrina and the aftermath, as it should have been. One of the other stories of the week came from New England, where the state of Massachusetts has proposed a plan to switch to open formats for all electronic documents created by state workers. A good summary is from ZDNet. If this plan is adopted, the state would switch to either PDF or Oasis, an open standard for documents.
I think the goals of this plan are excellent and admirable: documents created by, and used by, employees of the government should not be stored in a format controlled only by by a single company, which retains absolute control over that format. But in the way that governments seem to do, the folks behind the Mass plan may have come too late to the party.
Three months ago, Microsoft announced that Office 12 (the next version of Microsoft Office) will use XML as it’s standard document format. The flavor of XML they use will be “open and royalty free”. In other words, MS is responding to the same competitive pressures everyone else in the industry is facing.
Last week, Brian Jones of the Office team at Microsoft responds to the Massachusetts announcement with some interesting thoughts. To me, the most interesting line is this: “We are going to provide bulk upgrade converters that allow you to easily convert from the binary formats into the XML formats.” That’s great. Combine that with the commitment to provide XML reading/writing functionality for Office 2000 and Office XP, and it looks as though MS is opening up, at least a bit.
The reason that this declaration is not as meaningful as it would have been two years ago is that we are moving into a world where paper documents don’t mean as much as they used to. Microsoft Office, especially Word and Excel, are designed for one thing: producing paper documents. What happens when we don’t rely on paper documents?
When Microsoft announced plans for Windows95, they said that one of the big features was going to be a new file system that would allow you to organize your files (and, therefore, your data) in new, intuitive, powerful ways. For example, you could easily find all your files which contain information about 3Q Financial Statements, or pictures sent to you by your friend John W. The concept of directories would rapidly grow obsolete.
At least a year before Windows95 shipped (June 1995), Microsoft put the the plans for a new file system on hold. Periodically, they have raised hopes for the new file system to appear in WindowsNT, WindowsXP, and Vista (aka Longhorn). Each time, they’ve had to subsequently announce that it wasn’t going to happen.
Less than a year ago, when Microsoft announced that this new file system (called WinFS) was not going to ship in Longhorn, they got a lot of grief. These days, even average computer users are generating gigabytes of files. We need a better way to organize them than by directory and file name. That worked for DOS 3.1 in 1984. Surely we can do better 20 years later?
Tonight, Microsoft announced the first public beta of WinFS. They are not promising that it will be ready to ship when Vista ships (fall of 2006), but the fact that it’s in beta now is a good sign. Interestingly enough, they are also promising that it will run on WindowsXP, which means we won’t have to upgrade all of our machines to Vista to take advantage of this very long-awaited technology.
Here’s the official WinFS blog, and a video interview with the WinFS team, including a demo.
The New York Times has a story today about the removal of the final pieces of the automated baggage handling system at Denver International Airport. Within the next few weeks, the airport will be entirely using a manual baggage movement system, just like every other airport in the country. It’s a cautionary tale for all of us whole enjoy using technology and think that technology (of whatever flavor) holds great promise.
My favorite quote is from a professor who has studied the sytem:
"It wasn't the technology per se, it was a misplaced faith in it," said Richard de Neufville, a professor of civil and environmental engineering and engineering systems at the Massachusetts Institute of Technology. Professor de Neufville said the builders had imagined that their creation would work well even at the busiest boundaries of its capacity. That left no room for the errors and inefficiencies that are inevitable in a complex enterprise.
"The main culprit was hubris," he said.
It seems like the main culprit is usually hubris. Or greed.
I’m a regular reader of Robert Scoble, the Microsoft uber-blogger. He’s smart, articulate, and somehow manages to humanize the world’s largest software company. This week he wrote about Pandora, a new music service. He was so enthusiastic that I had to give it a look. I applied for an invitation and within a few hours I had one. So I tried it out, and discovered that the Scoble is totally right: this thing ROCKS.
It meets my two criteria for being really cool: 1) It’s something I want to start using right now (I’m listening to “George Winston Radio” while I write this), and 2) I can see it as part of a paradigm shift in the world. Here’s how:
People used to listen to radio, in part, to discover new music. That’s pretty much over, since Clear Channel programs songs that are designed not to be excellent, but not to suck. So how do we get exposed to new artists? Pandora will choose them for us, based on what we already like. Yet another reason to stop listening to radio.
You can sign up for a beta version of Pandora now, and hope to get an invitation. Or, next week Pandora will launch officially and you can buy a subscription for $36 per year. That sounds like cheap money to be able to hear music I really like from artists I’ve never known.
Several years ago, right before Christmas, I got a call from a potential client (we'll call him Al). Al had a screwed up installation of Windows Server 2000, on a brand new Dell server. His previous IT guy, a one-man shop, had gotten busy on other things and had left Al with a broken server. After meeting with Al and checking out his server, I gave him two options: 1) We can spend 10-12 hours fixing the Windows server, which had become terribly corrupt. He would get to use the software he paid for, and we would not have to rebuild his system. Or,
2) We can load the server with Linux, which will take 2 - 4 hours. We'd abandon the money he already invested in Windows, but he'd get a rock solid server which would do the two things he needed: share files and share printers.
It wasn't a hard choice: We installed Linux. Al was thrilled, but not with Linux. He was thrilled that he had a trouble free place to put files. His new, rebuilt server never acted up, never refused to let him login, never crashed, and never interfered with the way he wants to run his business. It wasn't Linux he was pleased with, it was what he could do with it.
Two plus years later, Al called me one day to tell me that he wanted to start using ACT! 2005 in his company. I explained that would require abandoning Linux, since this version of ACT! requires Microsoft SQL Server. He was frustrated: "But I really like Linux!"
Yep, that's the SMB relationship with Linux: When you need to do certain functions, like share files, share printers, and run web sites, Linux is great. It's rock solid stable, cheap, and doesn't demand the latest, greatest, most expensive hardware. But Microsoft has, by design and as a consequence of being the world's largest software company, created a whole ecosystem of products which demand Windows or Windows-dependant services (like SQL Server).
So as an SMB right now, you're stuck using Windows Server 2003 (complex, expensive, and difficult to maintain) or Linux (stable, but doesn't run any of the useful MS programs like Exchange or SQL Server).
Are there any other options? I think so, but they don't involve "competitors" to Linux or Microsoft. They involve thinking about the problem in an entirely different way. More to come.
In the early days of networks (1985 - 1991), the technology was designed for small groups of people who wanted to share files without using floppies. By the time Novell 3.11 rolled out (1991), it was stable enough that even larger companies wanted it. Microsoft entered the game (with LAN Manager) about that time, and replaced it with Windows NT a few years later. In 1996 Microsoft introduced NT 4.0, and within a few years they absolutely owned the SMB networking market. No one installing a new network installed Novell, and most organizations replaced Netware with NT.
But, Microsoft looked at the corporate market (businesses with more than 500 users) and saw that Sun and IBM had that market. So Microsoft began designing their products to compete for that size customer. Windows 2000 was the start in that direction, and Windows 2003 Server completes the shift from a focus on small organizations to a focus on very large organizations. Put simply, Windows Server 2003 is designed for businesses with more than 100 users.
The problem with Windows Server 2003 is that it's designed to satisfy the needs of ExxonMobil and other companies with thousands of users. It is built around Active Directory, which can perform all sorts of functions for zillions of users in forests, trees, organizational units, etc. But here's the kicker: For companies with less than 100 users, most of that power is actually a hindrance, rather than a help. Large organizations need a way, for example to replicate directory information reliably and efficiently across many servers. SMBs need a way to keep this information on a single server or, at most, two servers. They don't need clustering, or volume shadow copies, or any of the other large company features.
The problem with Active Directory (and, be extension, all of Windows 2003 server) is that it can only achieve it's power through complexity. That complexity comes in two forms: code complexity and interface complexity. Code complexity means that Windows Server 2003 is, basically, a really complicated computer program. Unfortunately, really complicated computer programs have bugs and break down Interface complexity means that it has lots of buttons to push and knobs to turn. That makes it too difficult for most SMBs to manage.
BTW: It's been interesting to see the effect of Windows Small Business Server 2003. When they introduced it, Microsoft priced it very aggressively, to compete with Linux. But they didn't want to spend the money to design a product for SMBs. Instead, they just took Windows Server 2003, which is already complex, and packaged it with Exchange 2003, ISA server (don’t ask), and several other pieces of software. The result is a server that is way, way too complicated for most SMBs to manage on their own. Even with professional technology support, it's still a pain.
So, is Linux the answer to all this complexity? Not necessarily. I'll talk about that tomorrow.
Today Google announced that they’re getting into the instant messaging game, with a product called Google Talk. There are lots of stories about it, including a quick review from downloadsquad.
My take is that it doesn’t offer much (anything?) that you can’t get right now from rival products from MSN and Yahoo, so there is no need to switch if you already use one of the other services. However, if you are new to IM, this might be a good place to start. According to the reviews, Google has done their typically good job in design and execution. Google Talk is a LOT cleaner than MSN Messenger.
Longer-term, the most interesting part of the announcement is Google’s support of open standards. Google Talk is built on the Jabber/XMPP protocol, which is open source. There is a rumor that AOL will shortly open up their proprietary AIM system by including a gateway to Jabber. I’m hoping this might start a process whereby all of the major IM networks can talk to each other, like email systems do right now.
I’ve got colleagues / friends / relatives on all three major IM networks (AOL, Yahoo, MSN), and it started to be a real pain to run all their different clients. I finally gave up and am trying Trillian, which costs $25, but speaks to all of these networks, and Google Talk as well. It’s not perfect, but it’s a ton better than running four different IM clients.
It’s kind of amazing how important IM has become.
In our business, we regularly advise clients as they purchase new servers, and new server operating systems. Right now, it seems to me that the SMB market does not have good choices for server operating systems. We are forced to help our client choose between two not-very-good options. The first, Windows Server 2003 (sometimes in the Small Business flavor), is clearly not designed for them. It's a slick option, and marketed aggressively by Microsoft. But it's crammed full of too many features requested by the Fortune 500, and it lacks the stability that SMBs require.
The second option, Linux, was not designed with anyone in mind. It has amazing stability, but a user interface that only a geek (yes!) could love. I can't think of a SMB client we have that could administer a Linux server by him/her self. Or even make minor configuration changes.
In my next two posts, I'll try to articulate more completely what I see as the problems in each approach. Then, in a third post, I'll offer my thoughts on how both vendors and SMBs can best deal with the current mess.
At Gnomedex two weeks, I had the occasion to visit with Buzz Bruggeman. I’d known about him and his company for some time, through references on various blogs. And I’ve been interested in ActiveWords (his software) for a while now. The other day, I listened to the podcast of a speech given by Buzz, and it finally convinced me that I needed to download and try ActiveWords.
If you’re curious, it’s a utility that runs all the time on your computer, watching for certain words you type, and then acting on those words. For example, I can type “xl” from anywhere on my computer, then hit F8, and Microsoft Excel will come up. I don’t have to type my hands away from the keyboard, click the start menu, click the Programs menu item, etc. I just type “xl”, hit F8, and I’m working in Excel.
I’ve been trying it for three days, and my experience has changed on each of those days, so I’m not ready to review the product yet. I’ll get to that in a week or two. For now, what interests me is the “culture” of the software:
When I buy new software that I am going to use every day, I want some sort of relationship with the person or company that produces it. For example, I use FeedDemon as my RSS aggregator. Not only is it a great product, but I’ve exchanged email with Nick Bradbury, who wrote it, and I’ve posted questions on the FeedDemon support forum that get answered quickly (usually in 8 hours or less). Nick blogs, and in his blog he addresses his future plans for FeedDemon and issues of concern in the RSS community. Because of those interactions, I have confidence that the product will be updated and supported.
In contrast, the culture of ActiveWords is much less human, and gives me much less confidence. Although Buzz blogs regularly (and well), the other team members apparently do not, or at least I can’t find their blogs. That means I’m not hearing what they are thinking about, and where ActiveWords is going. And these days, a software package without a future is a serious concerns.
The second concern is the activity level, or lack thereof, on the ActiveWords user forums. Forums like this have become the standard way for software companies to provide support for their products. I am used to posting a question and getting a relatively quick answer, either from the company itself or from another, more knowledgeable user. I posted a simple question to the ActiveWords forum on Sunday night and almost three days later I haven’t gotten any replies. This was a relatively simple question (indeed, I figured out the answer a few minutes later), but I left it there just to see how the company and the community would respond. They didn’t, or at least not yet.
It’s interesting to me that my expectations for new software have changed. I no longer simply expect it to perform a useful function at a good price; I also want to have a positive, interactive relationship with the company that produces it. This is not a knock on the ActiveWords software, but it will certainly factor into my analysis of whether I want to use this product and recommend it to my clients.
Recent Comments