As you know if you’ve read this blog earlier this week, I was at Microsoft’s Search Champs shindig. If you want to know more about what went on, there are plenty of other bloggers doing brain dumps. Me, I just want to write about one thing.
I’ve been talking about widgets for a long time now; both the ones from Konfabulator (now Yahoo!) and the ones from Apple (and yes, I know that Microsoft calls them Gadgets, but I’m calling them all lower-case “w” widgets here to group them together). And what I love love love about them is the idea of lightweight applications that can be created by people who aren’t programmers, aren’t developers, don’t even think of themselves scripters — people who, if asked, might call themselves Web designers. Or just designers. Or sometimes Web developers, but when pressed, will volunteer that they aren’t “real programmers.”
They know HTML. They know CSS. They know JavaScript. And as people from these corporations keep pointing out, that’s all you really need to know to create widgets.
Where people from corporations who implement widgets keep screwing up is that they think that they have to then sell creating those widgets to their usual developer crowd — the ones deep into Cocoa or C# or whatever. And the result is a big honking silence. Well, yeah; if you know how to create “real” applications, why on earth would you want to cripple your capabilities? What isn’t happening is trying to sell creating widgets to their natural audience: those Web designers I referred to above.
This post ain’t the first time I’ve ranted about this. For instance, I did this rant straight in the face of Apple’s Dashboard Widget crew back at WWDC. I did it again to MS this week when they talked about Gadgets.
One thing, though, was very, very different between those two rants: the Microsoft folks listened. Or at least they did a darn good impression of it. I spent a bunch of time talking to Steve and Sanaz, and I think they really understood what I was talking about and what I was trying to get at. For instance, Sanaz told me that she’s going to be doing two sessions on Gadgets at Mix06 (Microsoft’s upcoming conference which they describe as “a 72-hour conversation between web developers, designers and business leaders”). I asked her if either was “Gadgets for Designers,” and she said no. “Well, why not?,” I asked. And it looked like she got it.
The actual proof will be in what Microsoft does next, of course. They might not care. They might care, but internal politics make it irrelevant. Or they could be a huge corporation that can’t turn its message around that quickly. But I’ll be paying attention to see, and that’s how I’ll know whether the trip to see them was worth my time.
Two bonus comments for those who’ve made it down this far:
1. Here’s Apple’s current Top 50 Dashboard downloads. How many of them consume Web services? How many are just fun static doohickeys?
2. I ran the Dashboard Developer BOF at Macworld Expo earlier this month.
- Number of Apple employees who attended: 1.
- Number of Microsoft programmers from the Redmond campus who work on Windows who attended: 3.
This does not surprise me at all – most Apple engineers do not get time off to go to the show, and many of them don’t go at all. I know this was an after-hours thing, but it would still mean leaving campus by 4:30 and fighting rush hour traffic to get there. Unless they already happened to be at the show and involved in widgetry, they most likely would not have gone.
It doesn’t negate your point, but I don’t think this number adds to it, either. The person at Apple you need to get on board with your ideas is a Web technologies evangelist, whomever that might be.
Dori,
I’m not really familiar with these widget or gadget things but it sounds great. I’m a programmer but only because I have to in order to accomplish the things I want to, not because I love programming itself.
I’ve been meaning to check out Ning, Marc Andreessen’s startup. I guess they are offering some sort of web apps builder with minimal programming.
Maybe building lightweight web apps with no programming is Web 3.0?
Meh.
They’re all modal. Gadgets only work in a browser, Widgets only really work in Dashboard.
They’re a neat idea, but the implementations both blow.
Anne
Take a look at this.
most Apple engineers do not get time off to go to the show, and many of them don’t go at all.
Then someone from Apple shouldn’t ask me to provide them with 5-6 passes, should they?
The person at Apple you need to get on board with your ideas is a Web technologies evangelist, whomever that might be.
That would be Alan Samuel, Apple’s Java, Dashboard & .MacSDK Evangelist (ref) — who’s the one who asked for those passes. If you want to try to talk to him about this stuff, I wish you all the best; I’ve had very little luck reaching him.
BTW, I’m not blaming Alan for not showing up to the BOF, as I understand he had an accident over the holidays and didn’t make it to the show at all. It’s the 3-4 other people he said he needed passes for where I wonder what happened to them.
And if anyone wonders why someone who writes books & articles about Java and Dashboard has trouble reaching the Java and Dashboard evangelist, all I can say is: you haven’t dealt with Apple much, have you?
Maybe building lightweight web apps with no programming is Web 3.0?
How about 4.0? Or 5.0? I like the idea, but I hate these silly numbering things.
Gadgets only work in a browser.
Not really. Okay, currently Gadgets only work in a browser, but that’s not the plan once Vista ships. Then there will also be desktop gadgets and sidebar gadgets. This isn’t anything NDA, btw; it’s been out there for months (see an outdated but mostly relevant overview here).
Widgets only really work in Dashboard.
The way you phrased this implies that you know that they work outside the Dashboard, but that you don’t like the implementation. That, you’ll have to take up with Apple.
They’re a neat idea, but the implementations both blow.
Offhand, I like Apple’s implementation better. I don’t think that it blows, just that it could stand some improvement. MS’s could use some serious improvement, and I gave them some suggestions as to how that could be done. As I said, it’ll be interesting to see how much (if any) of my suggestions will make it into something that ships.
Lisa: thank you for plugging my stuff for me!
I’m hard on implementations. They either work well, or they blow. Both gadgets and widgets have kinda craptacular implementations. I know you *can* run a widget outside of dashboard, but I’ve yet to see one that made me care enough.
Even the ones that are actually web services applications are kinda ‘eh’. I’ve just not seen anything that makes me care that much.
The number of web service widgets in the Top 50 list may mean that people prefer static ones (thus, they’re more popular) or that more people produce static ones.
I’m not sure why the number of web service-based widgets is important. People are going to have ideas for widgets. Many of those will not require a web service, because they’re intended for local data. Many people will have little interest in setting up a web service just so they can put out a widget that uses it. And they may not be interested in creating a widget for some preexisting web service that someone else hosts.
And maybe the thing about widgets is that, really, they just aren’t *that* interesting.