Scripting News for 8/29/2006

I’m still hoping Republicans for Cut and Run takes off as a community project. Let’s look for Republicans who have decided we’ve wasted enough lives and money on trying to turn Iraq into the United States. If you participate in the political discussion, let other people know we’re looking for Republicans who have or are likely to speak out on the war. It seems with the election approaching there will be more. 

Last night I went to the Oakland Coliseum to see the A’s beat the Red Sox, 9-0. 

Om Malik: “Today’s news of Google CEO Eric Schmidt joining the board of directors of Apple Computer portends potential headaches not just for Microsoft, but for anyone with digital media ambitions.” 

BBCriver.com got a nice mention on Rocketboom today. Thanks. These guys are champs. :-) 

Yesterday, perhaps in an awkard way, I asked a question, curious to know how Feedburner calculates the number of readers they report for a site like TechCrunch. Clearly the number of hits they’re getting for Mike’s feed must be going up, and perhaps the number of unique IP addresses is going up too. Do they count all the readers at Microsoft as 1 subscriber? Is that just the number of times the feed was requested per day? How then would they factor in people who read it more than once. My aggregator reads it 24 times a day, am I counted as 24 “readers?” Bloglines just reads it once for N people who subscribe, how do they factor that in? Just curious to know what the method is. Since no one has to sign up for anything they can’t directly count the number of subscribers.  

Mark Fletcher writes that Bloglines reports the number of subscribers each time it requests the feed.  

Lorenzo Viscanti: “Feedburner’s count is just an approximation.” 

Feedburner: “Subscribers is an approximate measure of the number of individuals currently subscribed to your feed.” 

Revisiting another question I asked clumsily, about restrictions on speech that you accept when you participate in an invitation-only event, I posted a comment on David Weinberger’s blog this evening that attempts to explain the issue in 1-2-3 fashion. To be clear, although the questions are about Foo Camp, it’s really more general. 

EJ Dionne: “By Election Day, how many Republican candidates will have come out against the Iraq war or distanced themselves from the administration’s policies?” 

Duncan Riley names “kvetch” the word of the day. 

kevetch.com is Derek Powazek’s weblog, since 1995.  

goooooooo…ooooooooooooogle. :-) 

Four years ago today: “I would love it if the source for MORE were released. I think it would be a humantarian contribution of the first order, but it’s not mine to make.” 

Hurts so good 

Without comment, here’s a recent post from Tim O’Reilly.

David [Weinberger] — I’m curious about the ambivalence you express here. It sounds to me like you’re trying to curry favor with the well-known critics of FOO (and obviously, you failed) rather than giving us real insight.

Yes, FOO Camp is exclusive in the sense that it’s invite-only, but if you’re honest, you’ll have to admit that it’s one of the most diverse technical conferences you’re ever likely to attend. We make a real effort to invite people from different technical communities, people who ought to know each other, but don’t. One of our goals is to create new synapses in the global brain, so to speak. This is way more diverse than an open-to-all comers event that draws from the same community. And it’s precisely because we limit attendance that we’re able to manage that creative mix. Do you bake a cake by using whatever amount of flour, eggs, sugar, and chocolate you happen to have in your kitchen? There’s a range of experiment, but in the end, you control the mix because you know that some combinations are essential for the cake to work at all.

This is not to say that an open-to-all comers event can’t also work (assuming you have either a narrow topic that doesn’t draw a huge crowd, or else a venue that can hold all comers, and a set of goals that will work for a very large event.) But it would be a very different kind of event, and for our purposes, much less successful.

Stop worrying about what Winer thinks. He’s perfectly happy to attend exclusive events, as long as he’s invited. He has a grudge against me, for reasons only he understands, and despite his many other virtues, on this subject, he just needs to be ignored.

There are lots of ways to do great events. FOO is one. Bar Camp is another (and we’re honored to be imitated, and in fact have invited some of the Bar Camp people this year to be sure that they learn as much as they can about how we do it.)

Tim O’Reilly

If only… 

One year ago today: “Good morning. It seems New Orleans has been spared the disaster. That’s cool, it’s a very nice city, and so far has been totally lucky. Knock wood, praise Murphy, seems the luck has held.”

8 responses to this post.

  1. New Orleans is sign that it is time to remember what happen if we forgot about the world. the world answers back with the smack on your face if we forget.

    but today we remember and always remember.

    Reply

  2. I ran a consumer complaint handling company. Rent-a-Kvetch, Inc, from 1981 to about 1995. As a New Yorker, it’s hard to imagine anyone hearing the word for the first time in a recent blog post. :>)

    Reply

  3. Posted by bobby orbach on August 29, 2006 at 12:27 pm

    i remember MORE
    hey I miss those Living Videotext guys :)
    bobby
    Kvetch is like a Shlep or farKlempt or Chutzpa or Shtick they are unique words that have unique meanings, however a Shiksa not from NY may not have heard of them

    Reply

  4. RE TechCrunch and FeedBurner. If you look the feed (http://feeds.feedburner.com/Techcrunch), each post’s content ends with a series of images sourced off the FeedBurnr server. The images look like text reading “Digg this,” “Add a comment,” “Email this,” etc. but are really images.

    Each image appears to have a unique URL, so that one item’s “Digg this” is sourced from a slightly different URL than another item’s “Digg this.”

    FeedBurner is probably using hit counts on an individual image as a proxy for total number of readers. So even if bloglines downloads the feed once and shares it with hundreds of people, each of those people would separately request the images contained in the item. As for people who share a particular IP, you would count multiple requests for a given image as multiple hits, even if from the same IP, since an individual user will be caching images.

    Reply

  5. Re: TechCrunch and FeedBurner. Bloglines (and I believe all other popular web-based aggregators) reports the number of Bloglines subscribers for each feed in the User Agent field when it fetches a given feed. From this, FeedBurner at least knows how many Bloglines subs to factor into a total subscriber count.

    Reply

  6. FeedBurner uses different techniques for different clients. See http://www.crunchnotes.com/?p=229 for a blog post by Mike Arrington showing the breakdown of his feed subscribers from a few months ago.

    Reply

  7. Feedburner’s count is just an approximation (see this page http://blogs.feedburner.com/help/stats/subscribers_explained/popup/ ).

    After reading their page still the number of readers itself makes no sense to me. How can they create a precise approximation? May be there are 20.000 readers from aMicrosoft IP (using the same aggregator, IE7 ;) ), while there is just one crazy user here in Italy refreshing the feed in his aggregator every 30 secs 24/365.

    I think of this number as useful for a comparation between _similar_ blogs, an Alexaholic diagram for feeds.

    Additionally different kinds of blogs (i.e. Techcrunch and a personal blog written by a student) will have different subscribers (corporate users sharing the same IP vs home users).

    Reply

  8. There are pages dating back to the mid-90s that explain that it is pretty much impossible to get an accurate count of how many people visit your website. Hits vs. Visits vs. Unique Users are all [often inaccurate] approximations on the Web.

    See http://www.limov.com/library/web-stats.lml if you want a fairly reccent treatise on the topic.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 60 other followers

%d bloggers like this: