FREE Subscription to Dr. Dobb’s Digest: Same Great Content, New Digital Edition
Site Archive (Complete)
Events
Email
Print
Reprint

add to:
Del.icio.us
Digg
Google
Furl
Slashdot
Y! MyWeb
Blink
September 15, 2005
PDC 2005: Day Two

Jerry Pournelle
Four hours into Day 2 of Microsoft's Professional Developers Conference and Jerry is already on information overload.

It has been another long day. The sessions began at 8:30 AM which meant I needed to be in the convention center at 8:00, which meant leaving Chaos Manor at 7:00, which meant getting up at 6:30. I substituted Durk and Sandy's Fast Blast for coffee. That works, but it's a bit drastic. As usual the Press Room was well supplied with healthy breakfast food and good coffee. The Ethernet facilities worked. The Press Seating in the main hall is right up front.

Alas, there are no tables in the main hall, and no electrical outlets near where the press sits. Yesterday my HP 1100 TabletPC, which I use for taking notes in OneNote, ran out of power halfway into the presentations. This was my fault: Lisabetta, the TabletPC, runs for an honest 4 hours with a new battery, but over the years the battery tends to be run down and charged and cycled continually, and that original battery now runs only about 2 hours. My remedy for this was to buy a new battery, which I put into Lisabetta when we're going on the road; otherwise I use the old battery since she's docked all the time at home. Alas, I had forgotten to change batteries for the first day. I didn't forget that today despite having to leave the house at Oh Dark Hundred.

It was definitely worth getting down there for the session. Most of the morning was spent on the upcoming Office 12, much of it presented by Office systems VP Steven Sinofsky, who is a real enthusiast. His enthusiasm is catching, and I was ready to drink the Kool-Aid. However, since this is a developer's conference, there wasn't much emphasis on user capabilities, and eventually I noticed that. Instead it was mostly about how designers and developers can work together to use the new fully integrated Office 12 suite to build exciting new systems that will deliver a rich compelling user experience. After the ninth iteration of that phrase I was ready to scream, but I will have to say they showed a lot of cool capabilities. They also showed the new FrontPage. I use FrontPage, and this wasn't the FrontPage I know. This is FrontPage on steroids to be used in building corporate web sites with Wiki and search and blog capabilities built in, and RSS feed capability built into Word and working with FrontPage.

It sounded wonderful. Then I realized it was all built around InfoPath and Share Point Server, neither of which are familiar to most users. I wondered if this had any relevance at all to the work I do. Later in the Press Room Peter Glaskowsky speculated that the model for Office 12 is to get designers and developers into big organizations, train them for months, then have them turn out huge projects using Office 12. It will all be spectacular. There will be about a thousand of such cases, and each instance will earn Microsoft about a million dollars. This will be wonderful, but what won't be is a million cases earning Microsoft a thousand, and certainly not 10 million earning Microsoft a hundred. In other words, Peter speculated, this isn't for the general user at all.

Just about then I was invited to a Microsoft-sponsored lunch-time panel entitled "The Final Frontier of User Experience." Participants were Paul Sidlo, Founder, CEO, and Creative Director of REZN8, winner of Emmy awards, and computer graphics expert; Steve Dadoly, Director of Development, Infragistics, "a software tools company," that generates coding tools; Bola Rotibi, Senior Analyst, Ovum, a company that analyzes and comments on high tech marketing trends; and Thomas Murphy, Group Program Manager in Microsoft's Visual Studio Team System group. Microsoft is a software developer and publisher and sponsor of this conference.

I listened for a while to lectures on how the experts know how to bring about satisfying compelling user experiences, and where to place various information on the screen, and how the goal is to generate emotional responses in the users, particularly to bring about joy! Joy! Joy! Then we had another lecture that included mention of Plato's Golden Mean. There was other esoterica. The experts have arrived, and when the designers, who understand these things, can work with the developers who know how to generate code, the millennium will truly be with us.

I muttered to myself that it was Aristotle, not Plato, who had several chapters of some of his best works devoted to the Golden Mean. Classical Greek sculptors and builders used that ratio a good bit, and it certainly wasn't discovered by either Plato or Aristotle, but Aristotle wrote more about it than Plato. Still, Plato was a Pythagorean, and did require mathematics for entry into his school, so we can let all that pass. Then came more esoterica. Somewhere in there I think I lost it.

When it came my turn to ask a question, I said that we'd heard all this before. At every launch of a new version of Windows, and every new upgrade of major applications software, we are told about the rich compelling satisfying user experiences. When it comes out it's still spinach. I told them that they ought to hope that the software generates no emotions at all, because what I mostly see when an emotion is generated by software is hatred; and that despite all the promises, the help files and instructions seem to be written by graduates of that secret school that teaches how to write up help and instructions so that you can prove that you covered the subject, but no one who didn't already know how to accomplish the task being looked up can possibly learn how to do it. My wife's phrase is "Clear only if previously known." Why, I said, should we believe that this time it will work?

The answer I got was basically that it's all different this time, because everything is so much more complex. This wasn't reassuring. I recalled Rocky and Bullwinkle. "Watch me pull a rabbit out of this hat." Rocky says "That trick never works." Bullwinkle looks smug and says "This time for sure!"

At which point one of the panelists offered to take me outside, where I suppose he intended to beat me senseless. I thought this appropriate: They don't want to listen to a user, even a fairly experienced user. Why would they want feedback from a mere user? That's a far change from the old Word team when Chris Peters used to listen to anyone who used Word in hopes of getting a way to improve it.

Rather than stay to be beaten senseless, I wrote a note to the panel and left. I don't know if they got the note, but it said: "Yesterday at the plenary session, the Microsoft VP who spoke about the future of Microsoft applications told us 'We find that when users ask for new features in Office and Windows, nine times out of ten the feature is already in the program, but they don't know how to find it.' I rest my case."

If they did get the note, I haven't had a reply.

Later, though, I wandered over to an area full of Office worker bees, the people who actually have to implement the programs, and I spent a fairly pleasant hour. They didn't know a lot about the esoterica, but they did like to talk to users, and they were acutely aware that most users can't find many of the features in Office. They hope to build more wizards, and make the help files more useful. My contribution to that conversation was that disk space is very cheap, and they have very good search engines: Instructions can be keyed to a lot of different key words and question subjects, so that users don't have to guess exactly which term is the one used by the developers. Or designers.

I left that session feeling considerably better. The consultants and designers and esoterica experts seems mostly concerned with making things look better, so that everyone thinks it's cool; when an actual user intrudes, their first thought is to take him out in the parking lot and beat him senseless. The people who actually implement the code already know that the users are tearing out their hair, and the emotions generated by the software tend to be rage, frustration, and hatred, not joy! Joy! Joy! So we will see. I guess it's the eternal struggle: the Enlightened telling us Benighted what's best for us, and Designers designing to a theory because they understand us and know what's best.

Or perhaps I am unduly cynical and stayed up too late last night working on this stuff.

The Server Scene

Microsoft Server 2003 is a significant improvement over Server 2000. Alas, I haven't installed 2003. Now there will be even more in Vista server. It's clear I need to update all my servers, most of which were built from ancient hardware to begin with.

In particular, I do want to try running server versions of Office 12, with some of the work sharing and tracking features implemented. Now that BYTE no longer has 32 editors in Peterborough, I have to be a lot more careful with my columns. The old BYTE had what was probably the best collection of experts in the world, all available at my need. The new BYTE and Dr. Dobb's Journal groups are darned good, but there are fewer of them. Consequently, I have to rely on some long-suffering friends who read advanced copies of my column and catch most of my silly errors.

From what I saw at PDC there are a lot of new tools that will make it easier for them, provided that I implement server editions of the Word software. Since some of my friends are Apple Users, and others run a Microsoft-free Open Software environment (mostly Linux), it will certainly test the versatility of Microsoft's new server software when I try to get all those people working cooperatively. Perhaps I can't do it at all; but it will be interesting to try. Once again, the key to much of this is InfoPath, and Access, with new Search tools. All this promises to be exciting, different, powerful, and secure, to assure a rich compelling satisfying user experience. Before I actually swallow that Kool-Aid I think I want to try the software, which means my first step is new server hardware. Two new servers, actually, because while I am at it, I want to install an Open Source server just for contrast....

LISP with a Human Face?

One of the more intriguing presentations was a demonstration of List Processing. After a while I realized I was witnessing what I called in my notes "Human Usable LISP."

Those new to the computer scene may not be familiar with LISP. LISP is a List Processing computer language written in the 1960s by John McCarthy now Emeritus Professor of Computer Science at Stanford. McCarthy founded the Stanford Artificial Intelligence Laboratories (SAIL), and he and MIT's Marvin Minsky worked closely together on many AI projects. The LISP language is fully recursive, and unlike most computer languages, not only is there no strong typing and range checking, the language does not distinguish between program statements and data. It's thus very easy to create self-modifying programs, which are a nightmare to commercial software developers, but precisely what AI investigators want.

None of this is relevant here, because the new list processing capabilities are not based on LISP and don't work like LISP; but they do have some very powerful features that remind me of capabilities in LISP, or did this morning when I was listening and making notes. List Processing capability will be built into Share Point and FrontPage, including capability to generate RSS Feeds from any list. Lists can be bound to calendars, or to any Access data base. We were treated to several examples of what you can accomplish with this, including generated new lists ordered and aggregated in any conceivable way, all done intuitively so that you aren't writing LISP statements and you don't have to be John McCarthy to figure out how to accomplish what you want to do.

There will be a new Access as well as FrontPage in the new Office 12, and it will all allow use of InfoPath as a means of input. This seemed more exciting to me when presented than later when I began to think maybe it wasn't for users so much as designers and developers. Being told they wanted to take me outside as response to my user feedback wasn't an encouraging sign. Of course a developer show isn't necessarily the place to learn just what new features are intended for users.

The New MSN Search

My last appointments for the day were with the MSN Search design team. As you may or may not know, MSN Search is available free for most applications and will include MSN Virtual Earth. There are some restrictions on commercial use, but even for commercial web sites there's a very generous policy for free use. Check the web site for more details.

MSN Search attempts to be more sophisticated than other web search engines. Google tends to use frequency of link algorithms. MSN search watches as you use it; if a lot of users do a search, then visit the top site on the list but don't stay, jump back, and go to a site lower on the found list, MSN search notices that and downgrades the quality of that site. After a while that site is downgraded, and it may fall again and again according to user patterns. This is definitely cool. I tend to use Google by habit, but lately I have tended to use both Google and MSN search, and I may yet become an MSN search convert. We'll see.

Universal Studios

Microsoft conferences usually include a "fun evening" at some event spot. Last WinHEC in Seattle it was a night at Paul Allen's Music History museum. This year's PDC in Los Angeles we had a night at Universal City. I live not 10 minutes from Universal City and I haven't been there since the last Microsoft conference in LA. It's still a fun place to go.

Tomorrow I have a conflict: There's PDC still going on, and also Marty Winston's Quick Launch at the Beverly Hilton. I long ago promised Marty I'd go to his party, so that's where I'll be. Tomorrow morning's PDC presentation is about servers, and I suspect I know about as much as I am able to learn at this stage. I'll try to get town to breakout sessions tomorrow afternoon.

As usual, by the time I've been four hours at a PDC I am on information overload. That's better than running on information empty.

Jerry Pournelle is science-fiction writer and DDJ columnist. Jerry can be contacted at jerry@jerrypournelle.com.

7

TOP 5 ARTICLES
No Top Articles.



MICROSITES
FEATURED TOPIC

ADDITIONAL TOPICS

INFO-LINK