From: Dave Winer (firstname.lastname@example.org)
Date: Tue Jul 25 2000 - 10:56:29 PDT
Sally says the proper etiquette is to post the full text to the list so
people reading offline can read it. That works for me, so here's my
contribution. Please excuse the confusion.. Dave
--- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- --- ---
--- --- --- --- --- --- --- --- --- --- ---
We're in the last stages of preparing a new product for market. It's the
next step in the development of Pike, which we released as a free public
beta in March of this year.
The software works. I'm using it now. It does all the things that Pike did,
but it does something new too.
Now, we can't call the software Pike, because there is already a scripting
language with that name, and scripting, while hidden from the end-user,
plays a big role in the product. I also think it's good to re-think the
name, because it forces me to think about what this product is today and
what it's likely to be in the future. A good name will make sense now and in
So what does this unnamed product do? It's a personal radio station. Our
software makes it easy for you to program music the way a DJ programs music
(it's probably easier and more powerful than professional radio station
software). Scroll through your music list, double-click on songs to add them
to the queue. In the background, your servant, our software, is choosing
songs from the queue and playing them.
You can also add a <i>list of songs</i> to the queue with a single menu
command. (The user interface is outlining.) You can have as many lists as
you want, but there is only one queue.
The software is also an HTTP server. You designate a folder that you want to
share. Any playlists you put in that folder are available over the Web. Not
the songs, <i>we're not enabling MP3 sharing</i> as Napster and others do.
Just the playlists.
Think about this as an Intranet application. Music in business. An untapped
market, but a huge one imho. (Hint to the music industry, businesses
generally pay for copyrighted material.)
And this is also a www application, because there's interesting information
in your playlist, especially when it's aggregated.
The aggregator is not online yet. When it opens you'll see our favorite
songs, literally, the songs that are most loved by people at UserLand. But
then we'll open it for everyone to register their favorite songs, it will
recalc once a day. The format it reads is a simple XML format, easily
produced by any scripting language on any platform. We want to know what
other people are listening to. So you can share more than your playlists,
you can contribute to a new rating system for music. Let's find out what
Internet users like to listen to.
(BTW, all the playlists are XML too. Registration is handled through
XML-RPC. Everywhere we've had a choice to be open or closed, we've chosen
open. Our software is subject to competition. The user's data is not locked
up in a proprietary format, quite the opposite.)
And of course it gets deeper, but let's not go there now. I'm still trying
to figure out how to name this thing!
Here's why that's such a problem. Because all that I described is just an
<i>application</i> of the product. A powerful application, for sure, and
developed in record time because the platform is so rich. The key innovation
is that all the power is hidden unless you want to see it. The software is
as easy to use as a Web browser or Napster. To an end-user it's a simple way
to edit a radio station, but to a developer, it can be much more, just lift
the hood. It is a platform, and somehow the product positioning, if not its
name, must reflect that.
And dare I say it, it's a platform for P2P applications, which is the rage
in Silicon Valley. We took a long-term view in 1997, and made a bet that the
market would swing to distributed applications. We invested where the
venture capitalists didn't want to invest. While they were selling
e-commerce apps to Wall Street, we were developing a powerful platform for
P2P. Now that won't mean too much to end-users, and it's risky because the
financial community is so flaky, and even though distributed apps are the
way to go, I may not want to latch onto the bandwagon.
Anyway, I have to come up with a name soon. The software is almost ready to
This archive was generated by hypermail 2b29 : Tue Jul 25 2000 - 11:05:27 PDT