Tags: futureofwebapps



The Future of Web Apps, day two

I’m feeling quite refreshed and ready for another day of geekery. There weren’t too many drinking shenanigans going on last night.

The official watering hole for the FOWA drinkipoos turned out to be a yuppie nightmare. The entrance hallway was filled with gaudy images that were probably intended to recall 1950s pin-ups but actually just looked like page 3 pages torn from a tatty copy of The Sun. The drinks were ludicrously overpriced and getting out of the toilets required a significant toll charge. All of this would have been mitigated if there were some ancillary benefits such as watching young nubile bodies gyrating on a dancefloor but a sign at the entrance made it very clear that dancing was forbidden. This being England, the sign added, “we apologise for the inconvenience.”

Before long, a rebellion was organised and a gaggle of geeks made a mass exodus to a lovely cosy pub across the street. Happiness and chattiness emerged. After that, there was time for one civilised nightcap in the hotel bar with the dynamic duo of Tara and Chris, Google’s Jonathan Rochelle (a scholar and a gentleman) and Natalie—free from Simon’s clutches while he worked frantically on his slides.

It’s day two of FOWA now and there’s still no sign of free WiFi. Khoi has kindly given me a BT Openzone scratch’n’sniff WiFi card he got yesterday so I’ll use that to dip in and out of the river of connectivity and expand on this running commentary throughout the day.

Mark Anders

Adobe kicked off the day with a Flex demo. Having attended Flash on the Beach, there wasn’t anything new for me here but it was interesting to watch other people’s reactions to the speed of Actionscript 3 and the ease of downloading an Apollo app.

Chris Wilson

Microsoft’s Chris Wilson is on stage giving a state of the Web address. He talked about the origins of Ajax, gave a nice shout out to microformats and he mentioned the power of tagging (Hi, Chris!). There’s plenty of talk about security which isn’t that enthralling to me personally but its probably the most important aspect of IE7 for most people on the planet. Alpha transparency in PNGs; now that’s more like it.

Khoi Vinh

Khoi is talking about The Future (capitalisation intentional) which will, as he says, be awesome. But first, let’s hear about some of the design challenges at The New York Times. He’s showing some nice examples of what art direction is. You’ll see art direction in the print version of the paper all the time, but the online counterparts are just templated. There are exceptions like the fifth anniversary of the September 11th attacks and the infographics for the November elections, but of course these are events that are predictable and can be planned for. For breaking news, real-time design just isn’t possible… yet.

Khoi makes an interesting point about the schizophrenia in new technology. At the same time that we’re getting into hi-def television and DVDs, we’re also flocking to YouTube even though the video quality is really lo-fi. And while SLR cameras are getting more and more powerful, we’re using crappy little camera phones more and more. This schizophrenia throws up some design challenges for a media outlet like The New York Times.

There’s no such thing as a free feature, says Khoi. And remember, the more expressive a designer gets, the more the user has to pay for it (download times and such). So for any new feature, there must be a really valid reason for it to exist. Oh, and options are obstructions. Too many prefs are a sign of unresolved design issues that couldn’t be squeezed into the main interface.

Thank you, Khoi. And now it’s Simon’s turn. Hmmmm… I wonder what he’ll be talking about: OpenID, perhaps?

Simon Willison

Oh man, Simon’s on a roll. Talking a mile a minute, getting jibes in at Microsoft, cracking jokes about Ben and Mena Trott… he’s got the audience in the palm of his twirling, whizzing hand.

Long story, short: OpenID rocks. If you’re creating any kind of membership-based site, you need to check this out. If you’re member of a lot of different sites, you need to check this out. Oh, and in case you missed it, both AOL and Digg announced support for OpenID over the past few days. The momentum looks unstoppable at this stage.

I love the fact that the evangelism for OpenID is coming from passionate developers like Simon, not from some corporate representative. Like the microformats movement, it’s bottom-up rather than top-down. Other companies are buying slots at this conference to pitch their products but Simon gets to talk about OpenID because it’s so freakin’ cool and can’t simply be ignored.

Ah, OpenID and microformats: now there’s a cool combo. Simon has won my heart and the hearts of everyone else in the audience, I suspect. He’s talking about portable social networks and everything. Bravo, Mr. Willison!

Jonathan Rochelle

After a pleasant lunch with some of the Last.fm posse, I’m back in the auditorium to hear what Jonathan from Google has to say about Google Docs and Spreadsheets (killer name, indeed). These aren’t the kind of Web apps I’m likely to use myself but I’m interesting in the technology behind them. I’m assuming that, given the complexity of the applications, the Ajax used will be of the non-Hijax variety.

Open Mic

Time to break out into something a little unusual. This, as Ryan puts it, is the user-generated part of the conference. Over the past few weeks, delegates have been able to log on to the FOWA site and vote for some short presentations they’d like to see at this point. The three highest-scoring subjects will now present.

  1. The virtual office. Okay, that works.

  2. A documentation technique called Jedi — Just Enough Documentation for Interactions. Great backronym!

  3. The topic with the most votes is… which apps will succeed and which will fail in 2007? Who knows?

Daniel Appelquist

And now it’s time for a talk on mobile. Let’s hear from Daniel Appelquist from Vodaphone. I’m not entirely sure that a provider is necessarily going to be the most subjective voice on this but we’ll see.

Actually, there’s something interesting stuff here, especially around the intersection of mobile and Ajax. There’s plenty of talk about standards, so that’s all good. I’ll have to corner him later for a chat.

Rasmus Lerdorf

Now let’s hear from the creator of PHP, Rasmus Lerdorf. He’s taking us on a trip down memory lane, looking at Mosaic and early versions of HTML and PHP. Rasmus basically wrote PHP to scratch his own itch—it’s the typical open source story.

Here’s a reassuring confession from someone who has written a programming language:

I hate programming. It’s tedious. It’s no fun. It’s like flying: sitting in a smelly metal tube with other people. But I love problem-solving.

Looking at PHP today, it’s a lot more verbose. The Computer Science geeks like it now but it sure has moved far away from being a quick and dirty tool for getting something done. Ironically, there are students today that only have a background in object-oriented programming and have to be taught what procedural programming is.

Here’s an interesting idea on why people join an open-source community: oxytocin, a neuropeptide otherwise known as nature’s trust hormone. That’s in addition to the usual incentives like self-interest and self-expression. It’s the same motivation that drives people to play World of Warcraft in a big way. Open source projects like PHP are like Web 2.0 community sites: Flickr, Digg and Wikipedia would be nothing without the user-contributed content. The same goes for any open-source project.

In addressing the issue of performance, Rasmus has lost me but that’s due to my own mental deficiency rather than any fault with his presentation style.

Security is even tougher. As he says, “basically, you can never click on a link.” He has two browsers: one for browsing and one for sites that have personal data. It’s kind of paranoid, it’s kind of sad but, when you understand the consequences of cross-site scripting, it’s entirely justified.

PHP5 makes it trivially easy to take XML from Web services and do stuff with it. I can vouch for that.

Time for a quick announcement.

Tariq Krim

Tariq is from Netvibes. I haven’t played with it myself but Mike Stenhouse was raving about it yesterday.

There’s a big announcement coming right now. Here it is… a Universal Widget API or UWA if you prefer a TLA.

If you care, you heard it here first folks.

Wait, here’s another announcement: support for OpenID. Yay! All the cool kids are doing it.

Right. Make way for the guys from Moo.

Richard Moross and Stefan Magdalinski

Print is dead? Bollocks says Richard. And of course he’s right. Derek Powazek would agree, I’m sure.

Moo cards are cool. I’ve got some: little cards with my Flickr food pictures and the URL of Principia Gastronomica. A significant proportion of this audience also have Moo cards. Best of all, anybody here can get free Moo cards if they give these guys a business card in return.

Business cards don’t have to be boring. They can tell a story.

With Moo cards, the difference makes all the difference. Y’know, Qoop launched much the same product—business cards made with the Flickr API—a week before Moo cards launched. But Moo could compete on the differences: unusual size and high-quality recycled card. Everybody talked about Moo cards; nobody talked about Qoop’s cards.

Partnership is everything for Moo. Without Flickr, they’d be nothing.

Marketing is a four letter word: free. Giving away free cards is great marketing. I concur: the free cards I got from Moo clinched the decision to buy cards from them.

The attention to detail in Moo’s physical package really seals the deal. There are little Easter eggs in there and the luggage-tag card that comes with every pack gets everyone talking. There’s an incredible amount that has to be done by hand but that’s what guarantees the right level of quality.

Now Stefan is giving a peak behind the curtain at the technical side of Moo. If you want to know what he’s saying, well, you should have come to the conference then, shouldn’t you? You can’t expect me to do everything now, can you?

The Future of Web Apps, day one

I’m spending more time in London than in Brighton this week. After BarCamp London 2 at the weekend I had one day to recover and now I’m back up for the Future of Web Apps conference.

Like last year, the event is being held in the salubrious surroundings of Kensington; normally the home turf of Sloane Rangers, now overrun by geeks. But the geeks here are generally of a different variety to those at BarCamp (although I’m seeing a lot of familiar faces from the weekend).

The emphasis of the conference this time is more on the business, rather than the techy side of things. It makes sense to focus the event this way, especially now that there’s a separate Future of Web Design conference in a few months. The thing is… I don’t have much of a head for business (to put it mildy) so a lot of the material isn’t really the kind of thing I’m interested in. That’s not to say that it isn’t objectively interesting but from my subjective viewpoint, words like “venture”, “investment” and “business model” tend to put me to sleep.

That said, the presentations today have been less soporific than I feared. There was some good geeky stuff from Werner Vogels of Amazon and Bradley Horowitz of Yahoo, as well as some plain-talkin’ community advice from Tara Hunt.

The big disappointment of the day has been WiFi. Despite the fact that Ryan paid £6,000—remember, he’s not afraid of announcing figures in public—nothin’s doin’. For all the kudos that BT deserve for hosting the second London BarCamp, they lose some karma points for this snafu.

The day ended with Kevin Rose giving the Digg annual report. He left time for some questions so I put this to him:

I see Digg as a technological success and a business success but I think it’s a social failure. That’s because when I read the comments attached to a story, people are behaving like assholes.

At this point, people started applauding. I was mortified! I wasn’t trying get in a cheap shot at Digg; I had a point to make. So after informing the crowd that there was nothing to applaud, I continued:

This is probably because of the sheer size of the community on Digg. Contrast this to something like Flickr where there are lots and lots of separate groups. My question is; should you be trying to deliberately fragment Digg?

The answer was a resounding “Yes!” and it’s something that he touched on his talk. Afterwards, I was talking to Daniel Burka and he reckoned that Digg could take a leaf out of Last.fm’s book. The guys from Last.fm had previously talked about all the great features they were able to roll out by mining the wealth of attention data that users are submitting every day. Digg has an equally rich vein of data; they just need to mine it.

Anyway, it was a good day all in all but I feel kind of bad for putting a sour note on the Digg presentation. Plenty of people told me “great question!” but I felt a bit ashamed for putting Kevin on the spot that way.

Still, it’s far preferable to make these points in meatspace. If I had just blogged my concerns, it would have been open to even more misinterpretation. That’s the great thing about conferences: regardless of whether the subject matter is my cup of tea or not, the opportunity to meet and chat with fellow geeks is worth the price of entry.

The comments of crowds

The Future Of Web Apps summit took place in San Francisco this week. By all accounts, it was an excellent two days although it did spark an interesting hand-wringing debate about diversity which reminded me of the best ever episode of Father Ted: “I hear you’re a racist now, Father”.

One of the speakers was Mike Davidson. During his talk about Newsvine and online communities, my ears started burning. Why, I do believe he’s talking about me!

It all goes back to this post I made where I talked about how crap most comments are:

I’d like to propose a corollary of Sturgeon’s Law for blogs: Comments should be disabled 90% of the time.

Mike made the point that he finds it frustrating not being able to comment on my posts. Fair enough. He also speculated that the lack of a comment facility here might well lead to a decrease in traffic. I think he’s probably right.

But here’s the thing: I’m okay with that. I don’t think lots of traffic is a goal to strive for. There’s no doubt that comments are a simple and effective way of driving traffic to your site, but to what end? Instead of having lots of visitors, I’d much rather have a small amount of the right kind of visitors.

I’ve tried to explain this to people in the past (especially people just starting out in blogging) but I keep running into the same problem over and over: nobody believes a word I’m saying. But I swear it’s true! I’ve seen the way that useless comments can lower the tone on other sites and I don’t want it happening here.

Let me reiterate that this problem is particularly troublesome on sites that cover a diverse range of topics. Narrowly focused sites tend to foster higher quality comments. That’s why I’ve got comments enabled on the DOM Scripting blog which is focused entirely on JavaScript, but not here on Adactio, which is a smorgasbord of any ol’ rubbish that pops into my head.

It’s definitely a challenge for a wide-ranging site like Newsvine which seems to be handling the situation quite well. It’s certainly doing a lot better job than Digg. The rude, pointless, spiteful bickering that goes on over there makes me want to block any referrals from that domain. Mind you, it could simply be a matter of numbers. Digg users have clearly left their Dunbar number in the dust while Newsvine still feels cosy enough.

I’ve been trying to get at the root of my issues with comments on blogs. Ironically, I was able to crystalize my thoughts through participating in the comments on a blog post by Bryan Veloso. Oh, the irony!

I realised that comments on blogs are trying to fulfil two roles. On the one hand, they are a feedback mechanism — “Good post!”, “Me too!”, “You’re full of crap!”, et cetera. On the other hand, people claim that comments are a great way of fostering conversation.

Well, which is it? Feedback or conversation? Comments are a so-so way of dealing with both although better tools exist. Email is better for feedback. Mailing lists, forums, and instant messaging are better for conversations.

Now that I’ve had my about the dual nature of comments, I can better address what I want from them.

Here at Adactio, I don’t want to start conversations. I’m not looking to foster a community. I already run one large online community and I’d rather keep this site separate from all that. I am, however, interested in getting occasional feedback or hearing what other people have to say about some of the things I write about here. So, after much deliberation, here’s the moment that almost nobody has been waiting for:

I’m opening up comments here… but with a twist. To encourage feedback whilst discouraging conversation, I’m turning to .

There are a number of factors that go into making a wise crowd:

  1. Numbers. Generally, the bigger the crowd, the better. I have no idea how many people read this blog so I have no clue as to whether there will be enough people to make this work.

  2. Diversity. A diverse range of backgrounds and opinions is vital. I suspect that my site is mostly read by geeks, but I know there are non-geek friends and family that also stop by. Everybody’s opinion is valuable.

  3. Independence. This is the clincher. To really get wisdom from a crowd, it is vital that each person is acting independently. For a practical demonstration, just think about the “ask the audience” part of Who Wants to be a Millionaire? The results are strikingly good because each audience member has no idea what the others are choosing.

Comments on blogs fall down on that last point. Traditionally, comments are visible, thereby influencing future comments. That’s good if you’re trying to stoke a conversation, but lousy for getting some honest feedback.

So here’s what Im going to do:

I will occasionally open up some posts for comments. You will be presented with the usual form: name, email, url, etc. I would greatly appreciate getting your opinion. However, your comment will not be published immediately.

Comments will remain open for a set period of time; sometimes a week, sometimes a month. At the end of this time, all the comments will be published at once. At this point, it will no longer be possible to add a comment.

I still need to iron out a few technical details. It would be nice if there were a cron job set up so that you could be notified when your comment goes live. But mostly it’s a pretty straightforward set-up. It’s really only a minor variation on the traditional comment model but I’m intrigued to see what the results turn out to be.

Like I said, I won’t be doing this for every post. I intend to stick to my rule of thumb and keep comments closed 90% of the time.

Let’s get the ball rolling. What do you think of this idea? How vehemently do you disagree with my assessment of comments on blogs? Exactly how pretentious and arrogant do you think I am?

Comments are open.