Coty's Weblog

Are RIAs trying to be too rich?

I was reading this post regarding Adobe Chief Software Architect Kevin Lynch’s comments at Fortune’s iMeme conference on a “fourth generation” of applications that would return to the desktop and enable offline access but would be built with HTML, JavasScript, Flash, and other standard Web tools. Reflecting on that statement reminded me of something that I’ve thought for some time: namely that too much effort is being spent on new programming models like AIR, Silverlight, etc. and not enough on desktop integration for standard web apps.

What do I mean by desktop integration? Well, a number of little things, really:

• Desktop icons
• Separate processes, comand/control-tab switching, etc.
• Drag-and-drop
• File associations
• Etc.

The applications that really point the way in my opinion are a number of the WebKit-based applications that have been built to access web-based applications like GMail (Mailplane), Campfire (Pyro), Pandora (PandoraBoy), and others. Mailplane is arguably the best example. Though it doesn’t have an off-line mode, it does have extensive desktop integration—including drag-and-drop file attachments, iPhoto integration, Growl notifications, etc. I’m a big fan of PandoraBoy, as well. It’s use of global hot keys for voting, skipping, etc. makes it possible to use in the background in ways that the standard Pandora app cannot be.

As nice as all those fancy desktop integrations are, though, I think what I like most about these applications are that they are separate processes from my browser. It is a small thing, but it makes a big difference. It means that if my browser crashes, they don’t, and vice versa. It means that I can switch directly to one of them with command-tab and back to my browser the same way. They let me get back to browsing in my browser and using applications for more specific tasks.

Of course, AIR does aspire to all this desktop integration. I haven’t followed Silverlight closely enough to comment. It just seems like there should be something simpler, more open, and more standard, though—some browser feature (and associated application descriptor) that would let a standard web app be “desktopized.” For the simplest web apps, it could be something like the Web Clip feature due in Leopard but creating applications rather than widgets.

There is a lot to consider regarding all this RIA and desktop integration stuff. Coté’s Update from RIA-land: AIR & Silverlight, What is an RIA?, Better GUIs, the Killer App is worth checking out. It was his del.iciou.us link that directed me to the post that inspired this. For more WebKit-based apps like Mailplane, as well as other related projects, check out WebKit on Rails.

Time's Decimal Point

You probably already know this, but I spun my wheels a bit because if it yesterday, so I thought I’d put it out there:

1. The Time class in Ruby stores Time values with microsecond precision.
2. MySQL stores datetime values with second precision.

Why does this matter? Well, suppose you have a test case comparing a time you set in your ActiveRecord with the time you get back from the database for that record, and you get this result:

  1) Failure:
test_reload_with_local(FooTest) [./test/unit/foo_test.rb:23]:
<Wed Dec 27 10:27:33 EST 2006> expected but was
<Wed Dec 27 10:27:33 EST 2006>.

That is a little confusing on its face. The values look the same. But, as The Pragmatic Programmer's Guide warns in its documentation of class Time:

All times are stored with some number of microseconds. Be aware of this fact when comparing times with each other---times that are apparently equal when displayed may be different when compared.

Unfortunately, I didn't read that until after the fact. The problem is clarified, though, if you to_f your Time values:

  1) Failure:
test_reload_with_local(FooTest) [./test/unit/foo_test.rb:23]:
<1167234684.26596> expected but was
<1167234684.0>.

The second value is what was returned from MySQL, which had truncated the fractional second component.

There you go. I don’t know whether this will happen to anyone else, but if it does, perhaps you will Google this and find it helpful.

BTW, the time and timestamp types in PostgreSQL support microsecond precision, so this should theoretically not be an issue there.

Time zones are fun, too, but that will have to wait.

Microsoft and Vocalocity in the news

spkydog beat me to it, but I thought I'd also point to Microsoft's announcement today that Speech Server 2007 will support VoiceXML.

This is exciting news for the VoiceXML community. It is particularly exciting for Vocalocity, as we licensed some of our technology and worked with Microsoft to develop their standards-compliant implementation of VoiceXML for Speech Server 2007.

More information and comments from both Microsoft and our CEO, Mike Dickerson, on the agreement can be found in our press release:

Vocalocity Announces Strategic Licensing Agreement with Microsoft

New gig

This has never been much of a work blog, but I thought I'd mention that last month I joined Vocalocity as VP of Engineering. It has only been a few weeks, but I'm having a really good time. I've got a lot to learn, but I've got great people all around me to learn from.

Certainly not the least of those is our CTO, Jeff Haynie, who got his blog going again late last year. His latest post addresses the role of of VoiceXML and CCXML in an IMS architecture and includes a link to his recent presentation on the topic at SpeechTek West.

I don't know how much the new gig will influence my blogging. I'll certainly leave the deep voice tech. to Haynie, but you may notice more posting on Java, software development, etc. here than there has been recently.

It was going to be a really cool post, too

I was just about done with a relatively long post. I was using Performancing's Firefox blog posting extension and switched from the editing view to the markup view to see whether an HTML entity I'd typed in would carry over. It didn't. Nothing did. Crap.

Oh, well. It was a post about Mark Liberman's Old School post on Language Log commenting on Joel Spolsky's The Peril of Java Schools post and how it was cool for two otherwise unrelated blogs to come together like that. Maybe I'll try to resurrect my thoughts, but it probably isn't worth it.

Dabble is deep

Belying its name, I think Smallthought's Dabble online database is pretty deep. The most interesting bit is its focus on incremental development. There is really no point in my belaboring it in text, as you should just go check out the video from their demo at the November Lisp Users Group meeting. Oddly enough, Dabble is a Smalltalk app—not Lisp. But I guess there is some solidarity between Smalltalkers and Lispers.

I desperately want to see something like this displace the awful practice of e-mailing spreadsheets around for various ad hoc databases. I think that will take integration of something like the recently announced Simple Sharing Extensions for RSS and OPML as I think you can only displace spreadsheets completely when you have an option for disconnected use, which I believe naturally leads to synchronization.

Thanks to John Ludwig for the original pointer.

Imagine if Microsoft expended this kind of effort on real security

I don't know whether reading this whitepaper on "output content protection" in Windows Vista makes me want to laugh or cry more. So much engineering that could be directed into so many more profitable areas. It is sad. And it is funny, too, because I cannot be the only one who reads all of these grand plans and then thinks, "But I can just point a good camera and mic at the system and still get a good enough copy to bootleg." It is called the analog hole. As long as viewers/listeners remain analog receivers, it is pretty fundamental.

Ed Felten covers this in more depth in his Hollywood Controlling Parts of Windows Vista Design post on Freedom to Tinker. Ed touches on what bothers me about all this here:

Law-abiding people will be paying more for PCs, and doing less with them, because of the Hollywood-decreed micromanagement of graphics system design.

I wonder whether any industry groups in other countries see this as an opportunity to undermine Hollywood's dominance. They should.

Words of Wisdom from Adam Bosworth

I just finished listening to Adam Bosworth's keynote from the 2005 MySQL Users Conference. It is a good listen for the technical vision (Does that make sense?), but it was really these few relatively non-technical comments that I really thought might be most valuable:

• “I was amused to hear about people using Microsoft Project. I always viewed it when I was at Microsoft as this secret weapon that was used to stop everyone else from being able to compete.”


• “Anything that takes 4 years is probably not worth doing.” (Talking about time taken to develop XQL spec.)


• “I have never been able to get people within one company to agree on the schema of one thing.” (Responding to question about semantic web, etc.)

You are that star footballer!

The Commentator

The L. Wall/M. Fleury self-importance slider is priceless. Love the PairOn chair add as well.

Thanks to Mike for the link.

Happy April Fools' Day!

Oracle, Oblix, Identity, and Integration

So, Oracle bought identity management vendor Oblix recently. I have to admit that I didn't get the significance of the acquisition at first. Then I read Oblix to help Oracle, PeopleSoft tools 'coexist' and realized what should have been obvious to me: it was about integration. Oracle previously acquired PeopleSoft—and JD Edwards along with it. Now, I'm sure there is a grand plan of unification—there always is. However, I'm also sure that it will take some time to deliver—if it ever does.

Given that, Oblix gives Oracle the ability to at least eliminate one of the single most annoying "features" of running multiple disparate application suites from a user perspective: multiple sign-on. Eliminating multiple authentication/authorization systems can also significantly reduce administrative and support costs.

This should have all been evident to me, as my old company acquired more than its fair share of overlapping application suites while I was there—and the integration was not easy. (That is something of an understatement/misstatement, but I'm not going into that now.) At my new company, I'm working to integrate applications from separate vendors using federated identity as a cornerstone. Agreeing on identity goes a long way toward integration, and federated identity offers some good patterns for loose integration.