DISQUS

For my own blogs (blog.gonze.com, soupgreens.com) I have no good way for people to return to old posts, so writing is completely transient. This can't be the right thing, I know. The blockker is that common widgets like a calendar-oriented browser for the archives don't fix the problem.

Maybe the Wordle approach is a good tool. I'd be interested in hearing about your experiences implementing it.

Ben: yup, it started as Foxytunes Planet.

Servers just aren't that expensive. There's no way they would account for costs at the level TC is speculating about.

I wonder about the direction of causation. To some extent I think that people who wanted the Iraq war found ways to believe whatever was necessary to justify it.

In fact, I'll bet that plenty of people who read your post don't agree that evidence of links between Iraq and al Qaeda has not been found, weapons of mass destruction have not been found in Iraq, and that world public opinion didn't favor the US going to war with Iraq. They'll think that you're either lying or willfully mistaken.

Among people who liked the Iraq war and are too literate to fake their own reality, I think a lot of them privately disbelieve in the facts.

It is mind boggling how passive that is. "The train isn't coming towards me that fast, so why should I get off the track if it might still stop?"

What an insanely great piece of work. The simplicity is really clever.

It's not really a question of whether extensions are a business but of how the mechanics of extensions fit your business objectives.

Client-side software has fewer users but more engagement and stickiness. The users you do have either spend a lot more time with your software or are much more committed to it. So if you're a business living in a web site, you can grow your (average but not median) engagement and lower your churn by growing features on the client side.

Or at least that's how it should work. There are obviously would-be businesses with client-side products that have low time spent and high churn.

I still miss my dog Elvis. That handle seems to be always taken, though.

comparing YMP approach

You walk it like you talk it, Mark. Good shit.

I just set up a dreamhost account to migrate my blog over from wordpress.com, cause I can't have goose at toni's site.

Fred, I love that the auto-playlisting feature of the player is what led you to this song. It totally validates our idea that all the media in the page would be analogous to a webcast stream or long-form podcast.

Q'd up on Netflix.

Thanks for the reco.

For all that licensed content they probably don't have international deals, and even if they did the ads they could sell would probably only cover the domestic market. Given that the royalties must be brutal, I'm not surprised.

Right on, Steve. It's easy to unfollow the overposters, so why complain?

I kind of like emayoh's Hype Machine bot posts, actually. I check out the songs pretty often.

$36 CPM is very very rich. Even if they get that on some page views, they don't have 100% sell through at that rate.

So, for example, no salad bowls made out of bacon on Mondays?

Seriously, "Supersize Me" had a definite impact on my eating fast food, and I know that that Robert Pollan's writing has a comparable effect for a lot of people WRT meat. He seems to have a common sense way of making the case for eating habits that are otherwise a bit new agey. Haven't read him yet myself, though.

Yup.

Someday the labels will learn that all they have to do is own the canonical URL, and then they can insert ads as needed.

See also:
http://gonze.com/weblog/story/7-11-4
http://gonze.com/weblog/story/7-12-4

Added to the list of YMP mashups here: http://yahoomediaplayer.wikia.com/wiki/Mashups_...

Working fine for me with FF 2.0.12 on OS X.

I strongly agree with this perspective. It's good to see you articulating and documenting it.

Sometimes the users just don't know what they want, and they need the product design to steer them in directions they didn't even know existed. But aside from that one exception, the rule is that developers have to be humble and view their job as watching and reacting to real-world usage.

The killer usecase for Webjay.org, for example, turned out to be exporting XSPF for third party XSPF widgets to render. This was very different from what I expected when I started on either Webjay or XSPF.

Hi Fred,

FYI, the new build of the music player with the improvements I mentioned is now up. See http://developer.yahoo.net/blog/archives/2008/0... for details.

thanks.

-Lucas Gonze

What makes you think that webmail isn't profitable, Robert? That doesn't make sense. It has rock bottom content costs and sky high engagement.

The only overhead is for coders, hosting and storage, which are a non-negotiable part of almost all web apps.

Fred, I think you'll dig the player. If you want more precise control over the player than with the default view, check out: http://yahoomediaplayer.wikia.com/wiki/How_To_Link

This is very much a first release and there are still bugs big enough to matter. For example the play state sometimes gets out of sync and you end up with two songs going at once.

Also, we're eating too much screen real estate in the default layout. In the next rev we'll have the same footprint as PlayTagger.

BTW, I am both the product lead for this player (I report to Ian) and Sylvia's peer at Xiph. The business and standards issues are aligned.

Oh yeah. Oh yeah. Steve, I'm blowing off Thanksgiving dinner and heading over for the slice.

bon appetit.