How Microsoft is quietly leaving its past behind

Since starting my new job over three months ago, I’ve been too busy to write anything here. I’m not complaining; I’m really happy to have this job and the opportunity it gives me to make “computing” better for people at large. Thanks to Canada Day’s being on a Tuesday this year, I and many of my colleagues are taking Monday as a vacation day to create a four-day weekend, so here I am, writing again.

I put “computing” in quotes because these days it’s harder than ever to know what to call it. I like that people often use phrases like “being on the Internet” and “going online”. Sure, you can use a computer without being connected to anything or anyone else, but the range of what you can do is so much more limited.

I sometimes talk about “personal computing” as a period of temporary insanity. (Apparently I haven’t written that in a blog post before now, although I did write it a couple of years ago in reply to a comment here.) We’re recovering from it now, and I find Microsoft’s way of dealing with the shift to be quite fascinating. Microsoft stands to lose more than anyone else, but isn’t silly enough to pretend that that things aren’t changing. Its response, Live Mesh, takes Microsoft into the new era, but in an old-fashioned way. Let me explain what I mean...

There doesn’t seem to be a generally accepted term for the emerging new way of using computers. Out there in “the cloud” of the Internet there are servers that have all kinds of applications, data, storage space, you name it. Meanwhile people use various devices, such as PCs and cellphones, that can connect to the Internet and therefore to all those services as well as to each other. What is now available to us is the powerful combination of all of these elements working together.

The advent of PCs is why Microsoft was founded, and selling software to run on those PCs is still how it largely makes its money. As things evolve, however, the need for that software is disappearing: using a PC operating system other than Windows no longer causes much difficulty to the user (I now use Mac OS at work, and not long ago I wouldn’t have been willing to), and even though I have Microsoft Office at both home and work I rarely use it any more since I’d rather use server-based systems such as wikis. To me, a device is just a way of getting at my “stuff” – which I really don’t want to be stored on a device that might get lost or stolen, or whose hard drive might (will, someday) fail, or might get screwed up by the actions of its owner (me) or by all those software updates that I don’t understand and just hope will work as they’re supposed to.

UPDATE: Somehow I forgot to mention this when I wrote this post yesterday: the other problem with being device-centric is that a lot of things aren’t generated by you on your devices. Like all the emails that people send you! And the instant messages, and all the news you read on the Web, etc., etc.

Well, the way that Microsoft describes Live Mesh is very device-centric, and the word “mesh” seems that way to me as well, even though as far as I can tell it is a comprehensive system that could be a solid platform for the future. “Devices are how we interact in this new ‘web connected’ world”, says the Introducing Live Mesh blog post; “however, ... it becomes increasingly difficult to keep the people, information and applications we depend on in sync”. The result is a focus on synchronizing devices with each other. But how is this actually done? By synchronizing them with “folders” – that happen to be stored on a server, not on any of your devices. In fact, “Live Desktop enables you to easily access your mesh anytime, anywhere, using only a Web browser.” You don’t actually need to use any of your own devices; any Web browser anywhere will do.

Even at Microsoft, the device is becoming just an access point to where things really live, out there in the cloud. It’s just that Microsoft doesn’t want to admit it.

Putting my income where my mouth is

As I blogged a few months ago, I recommend that you store your email not on your own desktop/laptop computer, but rather on a server that you can access over the Internet, and that if you prefer to do your email using an email program (like Outlook Express or Apple Mail or Thunderbird) rather than through a Web browser, you should use IMAP on an email service that supports it.

Well, on Monday I started working at Tucows as Director, Tucows Email Service.

Some of you may remember Tucows as the original software download site on the Internet; I used it as far back as 1994. Today, Tucows the company still runs that site but also does a number of other things. Its largest business is now to provide services for resale by over 7,000 ISPs and web hosting companies. The largest reseller service is domain registration; Tucows is actually one of the world’s largest domain registrars! Also in the area of domains is SSL certificates: your website will need one of those if it’s going to provide https:// access, e.g. if people can buy things by giving a credit-card number.

Then there’s the Tucows Email Service, which is my main focus. ISPs and web hosting companies need to provide email to their customers, but doing it themselves means having to keep an email system working reliably around the clock, with spam filtering and antivirus protection. The most recent Email Metrics Report from the Messaging Anti-Abuse Working Group, of which Tucows is a supporting member, showed 86.7% of all email as “abusive”, meaning spam. You may think that the spam that’s directed at you largely gets diverted to your spam folder, but you’re not even seeing the additional email that is so obviously junk that it doesn’t even get put in your spam folder! The spammers are always coming up with new tricks, and the total challenge of running an email service well means that it’s an activity best left to those who specialize in it. Even Bell Sympatico, Canada’s largest ISP (I co-founded Sympatico back in 1995), which had a perfectly decent email system, upon forming a partnership agreement with Microsoft chose to migrate to a Microsoft-hosted email system for the future. So the Tucows Email Service is an excellent way for ISPs and web hosting companies to provide quality email to their customers, complete with 99.99% uptime guarantee. So far it has resellers on three continents; I’d aim for all seven except that I don’t think Antarctica has an Internet industry.

My other focus is the fairly new Personal Names Service, a unique offering that’s built on top of the Tucows Email Service and also makes use of Tucows’ specialty in domains. Suppose your name is Yvonne Desjardins and you have a typical email address like (slightly misspelled to foil spammers) yvonne.desiardins@sympatico.ca or yvonned@hotnail.com. What if your ISP contacts you and tells you that you can have yvonne@desjardins.net for a small fee per year? You might well be interested. That’s what the Personal Names Service does, via its collection of about 40,000 domains such as desjardins.net. (It doesn’t own anything for the surname Jayasekera – I guess my surname isn’t all that popular in certain countries!)

One of the interesting opportunities here is to make Web-based email access even better than using an email program installed on your computer. Webmail always used to be the ghetto version of email; you’d use it only because your email service was free and accessible only though webmail (and not very nice webmail either), or because you didn’t have your computer with you. Gmail was the first large-scale email system that aimed to provide a high-quality web interface, but as much as I admire what they’ve done, the fact is that I’ve never used my Gmail account very much. As I wrote in that earlier post, I’ve always preferred to use an email program, so if even I can be converted to preferring webmail that will be a good sign.

I’m very excited about this new job. When I first used email about 35 years ago, it was much better than it is now: if I sent an email, I could be sure that it would reach its recipients (unless they were avoiding their email of course!), and there was nothing annoying about email either. Plus I could retract emails that I had already sent. Now, email is more important than ever, but it’s deteriorated. I will do what I can to help make email great again.

Mozilla Prism

In my last post I talked about how you can now properly access Gmail not only through its Web interface but also through a PC-based email program like Outlook Express or Thunderbird.

That’s an example of where a server-based application is accessed other than through a Web page. In this vein I’d like to mention Prism, which was announced a few days ago by Mozilla Labs and works with the Firefox browser. Prism (formerly WebRunner) makes Web-based applications behave more like PC applications and less like Web pages. Here’s an example:

I use a Web-based to-do list system called Vitalist, and I use it frequently throughout the day. But with lots of browser windows/tabs open at the same time, I may have to hunt a bit to find the one that has Vitalist. Not any more. A few days ago I installed Prism and told it to create an application with name “Vitalist” and URL “http://my.vitalist.com/”. Now I have what seems to be a regular Windows application:  there’s an icon on my desktop labelled Vitalist (I could also put it into the Windows Start menu if I wanted to, or the Quick Launch bar), and when I run it the taskbar shows “Vitalist” in the same way that it would show a traditional application like Outlook Express, completely separately from any regular browser windows I may have. I can navigate to it in the same way I would any other Windows application. Furthermore, the window doesn’t waste space with browser buttons like Back and Forward, nor with a location bar, because with Prism those are optional: it gave me checkboxes for them and I didn’t check those off. Although the window is actually a browser window, you’d never know it.

I’ve used Prism for a few days and although it’s an “early prototype” it works fine for me, and I love it. If you use any web-based applications a lot, like Gmail or Facebook, you may like it too.

So far it’s available only for Windows, but Mac OS and GNU/Linux versions should be available soon. More information, and a link to download it are at mozilla.com, specifically here.

One fewer reason to store data on your computer

In my post Online storage I wrote about why you shouldn’t store your data on your desktop/laptop computer, but instead store it on servers that you can access over the Internet.

That includes your email. Web-based email is popular, but what if (like me) you prefer to do your email through an email program (like Outlook Express or Thunderbird) than through a Web browser? Up to now you’ve usually had to download your email to your local computer, while (optionally) leaving your original copies on the server. While that works, unfortunately the communication goes only one way: when you read an email message through your email program it gets marked as “read” locally, but not on the server. Same for deleting messages and other things. And properly organizing the messages you send is usually a hassle.

There is a solution: it’s called IMAP (Internet Message Access Protocol) and is one way that an email program and a mail server can communicate. With IMAP everything is synchronized: whatever happens at one end gets reflected on the other. And the mail server is the “primary residence” of your email; you may have copies of various messages locally, for speed and for availability when you have no Internet connection, but IMAP meets my objective of storing your data on servers intended for that purpose.

IMAP isn’t new; in fact it’s older than the public Internet. But the popular free email services such as Hotmail and Yahoo! Mail and Gmail haven’t supported it, unfortunately.

This has now changed. Gmail has just added IMAP access as a new feature (though Google says it will take a few days to be rolled out to all Gmail users).

Since Gmail is so popular, its addition of IMAP means that a lot more people will no longer have any good reason to store their data on their local computer. The effect will be magnified if Gmail's competitors try to keep up by also adding IMAP.

Facebook as portal

I haven't posted for a while; too busy with work and things like InteractionCampToronto and the great mesh conference here in Toronto last week.

About three months ago I wrote about my experience with Facebook. My admiration has only grown with the addition of the Facebook Platform, and enormously so. It's been less than two weeks since it launched, yet my Facebook friends are busy finding apps to add. And when Facebook lets me know that a friend has added an app I haven't yet heard about, I always check it out to see whether it's of interest to me too. This is only the beginning of this new platform, a platform that while "new" is built on top of existing social networks, giving apps a better shot at viral spread and often eliminating critical-mass challenges.

I highly recommend a really great guest post on Techcrunch: The New Portals: It's the Bread, Not the Peanut Butter by David Sacks. It's about portals, and in particular talks a lot about Facebook as a new kind of portal. Even though I myself co-founded Canada's most popular portal, I largely lost interest in it a long time ago when the portal sphere started to stagnate, and later the advent of Web 2.0 massively changed the environment within which the very concept of a portal is applied. A good Web 2.0 portal doesn't have that much in common with a good Web 1.0 portal, and perhaps it's time to introduce a new term. "Portal 2.0" has been used to refer to personalized home pages like Netvibes and iGoogle, but to me that is too limiting: a portal is something that people use as a main jumping-off point to the Web, and Facebook now falls in that category. I invite suggestions for a new term in the comments.

Again, I highly recommend the post I linked to in the preceding paragraph.

Welcome to Web 2.0; wanna be my friend?

Unless you’re a complete hermit, you occasionally have to deal with strangers who want to befriend you for one reason only: so they can make money. Multi-level marketing is one source of that, but it’s been going on so long that you’ve likely developed an “is this multi-level marketing?” analyzer in your brain that starts up at the slightest hint and won’t stop until the question’s been answered to its satisfaction.

“Networking” is a more recent one, and unfortunately deciding between the good and the bad is often not so easy. In modern societies connections are much more fluid than in the past, and now that the era of the “permanent job” is ending, networking becomes an essential requirement for making a living. So networking isn’t so easily tuned out. (I’ve been a huge fan of Tom Peters for over 20 years, and it shook me to the core when I heard him say that everyone needed two skills and one of them was networking. I don’t even remember what the other one was, because it was something I decided I had and consequently didn’t need to worry about — but I knew that networking wasn’t my strong suit.)

Enter Web 2.0 and its “social networking”. In any social network, whether it’s a small interest group of people who like to knit scarves in the shape of a brontosaurus (“all brontosauruses are thin at one end, much much thicker in the middle, and then thin again at the far end”), or a huge ocean of pseudo-friendship like MySpace, the cost of reaching out to a fellow member is low. There have always been “joiners” who join every club they can find so they can meet more potential customers for insurance policies or whatever, but when meeting people requires actually going to a meeting it limits the amount of such activity. Online, reaching out promiscuously has little cost, and doing it in some quantity, perhaps even with a bit of personalization, is a trivial matter. Interest in email, a medium that’s largely meant to be person-to-person, is declining thanks partly to all the spam, and social networks similarly risk pollution levels that make the environment inhospitable. Women are at the forefront of this because of all the friend requests from men.

(I realize that there are many people who want as many online “friends” as they can get (“thanks for the add”). I’m not talking about them. I also think that their interest is a temporary phenomenon that will largely disappear once the novelty wears off and people realize that collecting 2000 “friends” is too easy to be worth any bragging rights.)

This post was prompted by a friend request I received from a fellow member of a social network who was looking to promote his new Web 2.0 venture. Not intrinsically a bad thing, but:
- he provided almost no information about what the product will do (“it has to be experienced”);
- all I can do now is to sign up to be included when the “alpha” starts up in future;
- he’s almost certainly never launched a product before (for instance, I’d be joining a beta, not an alpha); and
- it’s called Hypesphere. When hype is considered a good thing, count me out.

He’s not trying to get at my money, but he is trying to get at my time, and I only have so much. Request deleted.

Getting wikis filled in

“The Wiki field of dreams bothers me just because you build it doesn’t mean that people will do your boring content entry”
-Bryce Johnson (on Twitter)

I think you can build a wiki and get it populated, as long as you satisfy the following conditions:

1. Your wiki fills a need, one that’s not already filled. According to people who are in the wiki’s intended audience — not according to you or “management”.


2. The people who would use it include a high percentage of what I call analytical-retentive people, like computer geeks, librarians (hi, Connie), or policy wonks (hi, Mark).


3. You seed the wiki the way I’m about to talk about.
   

If you’re like most people, if you’re given a blank slate and are asked to put something on it, you’ll have a much harder time than if someone gives you a starting point that you can modify. Even an example of the kind of thing that’s desired constitutes such a starting point, e.g. if you’re asking someone for a description of a table and you want to know its height, materials, etc. you can give as an example a description of a bookcase that includes similar attributes. (Highly creative people do thrive on blank slates, but most people aren’t that creative, and furthermore the people you want populating your wiki are the ones who are more interested in knowing boring old facts than in being creative.)

If you just create a blank wiki, or have just minimal content in it, chances are high that nobody else will contribute anything. So you need to get the wiki started by creating a bunch of pages and putting something on each page.

That’s advice you’ll get from other people too, but I would add this: make those pages annoying. Annoying to people who are interested in the subject and are bothered by seeing it treated poorly, enough so that they’ll fix the problem. Analytical-retentive people are more bothered by flaws than other people, and furthermore are usually good at fixing them.

You can’t just create garbage as your starting point: you need to create something that’s going in the right direction, but is flawed. The better you know key people in your audience, the better a job you can do on this: ask yourself what kind of flaws would get those key people riled up and anxious to fix them. For example, if a wiki page is about how to use a Macintosh computer, you could seed it with some “information” that is obviously about Windows and is completely wrong for a Mac. This is where creativity can really come in handy: for seeding the wiki, not for populating it.

Misinformation is not the only way to seed with flawed content, but it can be an effective one. If you use misinformation, I recommend that your wiki be in a clearly stated beta mode until all the misinformation has been removed by users. (There is always the possibility that someone just removes misinformation without replacing it with something accurate, but it’s less likely to happen if you do your job well. There is an art to this.)

I’ve never seeded a new wiki by putting in provocative content, but I’ve successfully used the technique to seed individual pages in an existing wiki, and I find it powerful.