Scripting News

Evan Williams vs the Internet

Tue, 21 Jul 2009 16:28:28 GMT

In October 1994, at the dawn of blogging, I wrote a piece that actually shook the software world. At the time, the idea of a mere software developer expressing an opinion in public, unedited, in his own words, without the help of a major publication, was unheard of. It had never happened.

The piece was called Bill Gates vs the Internet. The thought was pretty simple. The tech industry was mired and exhausted. Too many BigCo's struggling to be the one who controls the future. As if a company could control the future. But the headlines in the business press encouraged them to think this way. Much as the leading tech blogs encourage Schmidt, Zuckerberg and Williams today to think of themselves as masters of the universe. They aren't and it's a losing strategy today as it was 15 years ago.

The problem for Bill Gates in 1994, the newly crowned King of Tech, was the Platform Without a Platform Vendor, the Internet. The difference between the Internet platform and the Microsoft platform was this: No Microsoft. No one to hold on to the family jewels. No one to put a developer out of business if they personally offended Bill. No one to keep the personalities of developers under control. No one to cut off their air supply.

In 1994, there was a revolution brewing. Bill didn't believe. But it happened anyway, even though he struggled mightily against it.

Blogging is one of the things that came out of this revolution, and along with it archives. So I can point to a piece I wrote in 1998 and it's still there. It was systematized, in software. This idea didn't come from a BigCo, and it didn't get killed by one. The free Internet solves problems pretty well. BigCo's don't solve problems.

So now instead of Bill Gates it's Evan Williams.

I read the piece on TechCrunch and thought it sounds like the transcripts of conversations from Microsoft in the mid-90s. Both were trying to compete with the Internet. Ev's problem is how is he going to keep his key engineers from defecting to the competition. How are they going to let developers use the "firehose" without using it to kill TwitterCorp. These are problems the Internet doesn't have. It doesn't employ any engineers, and when they leave one company to work for another they still work for the Internet. On the Internet no company owns all the data, so no one can control it. If you don't like the way a service works, use another.

The tech industry keeps having this argument with the Internet. It keeps thinking "this time we gotcha" but nahh, the Internet keeps right on going.

Moral of the story: If you find yourself in competition with the Internet, you should find a way out. Imho.

Frontier bug

Tue, 21 Jul 2009 19:25:47 GMT

Wanted to record this to be sure I get back to it some point.

Problem is with all HTTP requests emanating from tcp.httpClient.

Setting the timeout has no effect when its not possible to open a connection on the server because tcp.openStream doesn't take a timeout parameter. It's always 20 seconds, near as I can tell.

Once I get back into the C source again I'll have to check this out.

You can see the problem on the log page for rssCloud -- when testing the link back from a remote app registering a handler, the timeout is never less than 20 seconds, even when it's unreachable. I have the timeout param in tcp.httpClient set at 180 ticks (3 seconds), which is plenty to find out if there's anyone at the other end.

Dumb XML question

Tue, 21 Jul 2009 18:33:08 GMT

0. I know about View Source.

1. I use Firefox.

2. When I view my site's RSS feed I want to see the XML, not a stylesheet rendering of the XML.

3. To be clear, I want to see the actual XML.

4. Is there some way to force the browser to do this?

5. Please no lectures on how this isn't the way it's supposed to work. TIA.

6. I know about View Source.

7. I know about View Source.

Update: This post seems to explain what's going on. Most browsers do funny stuff with RSS.

Speedbumps and a city's carbon footprint?

Mon, 20 Jul 2009 19:54:22 GMT

We have a mail list for the back channel at InBerkeley.com, and from time to time a question comes up that requires research. If the question is interesting, my first impulse is: Write It Up!

Now, this is the result of 12-plus years as a blogger. I know my community loves interesting questions, and we have an informal approach on Scripting News that I'd like to port to InBerkeley.com.

So, in that spirit -- here's a question posed by my colleague Mark Haas.

Do Berkeley's infamous speed bumps, traffic diverters and other traffic-related policies, like politically-motivated, too-low speed limits raise the city's carbon footprint?

We just need a qualified author. Anyone know any traffic engineers, or perhaps someone at the UC Berkeley Institute for Transportation Studies? Other experts?

Walter Cronkite's 'Cosmic Disaster' editorial

Mon, 20 Jul 2009 14:54:54 GMT

In this week's Rebooting The News podcast, I chose Walter Cronkite as our inspiration of the week.

What a 140-char message looks like in RSS

Sun, 19 Jul 2009 17:37:57 GMT

Here's a screen shot to contemplate.

There are two 140-character messages. Each illustrates features of the new shipwreck I hope to sink, to create a new coral reef for Twitter-like systems to grow on and around.

The first three items in each message are fairly obvious:

<description> holds the 140-character text.

<pubDate> is the timestamp, when the message was created.

<guid> is the identifier for the message, so a reader can tell if they've seen it before.

This makes it possible for the messages to be edited after publication, a common feature requests from writers using Twitter.

After that come optional elements.

<category> works like tags in apps like Flickr or YouTube. You specify them in a dialog, blanks separate them, you can create tags with blanks by putting them inside quotes.

<link> is used to point to web pages. No need to shorten the URLs because they don't take up space in the 140 characters.

<enclosure> is how you attach media objects to messages. Again, no need to shorten the URLs. And since the clients know the media type, they can show a preview, or embed a player.

These all use well-understood elements of RSS 2.0. Nothing new needed to be invented.

Craigslist is progress

Sun, 19 Jul 2009 20:41:10 GMT

I don't think I've ever written about Craigslist here.

Probably because I don't spend much time thinking about it, or worrying about it. But I know that some people do, for example Terry Gross, the host of NPR's Fresh Air. It comes up when people talk about the Internet destroying things that matter, like the classified ads in newspapers. At one point in an interview with Wired editor Chris Anderson she asks, in a bewildered way, what happened. She was saying it was a shame that Craigslist comes along and does what the newspapers were doing, for a fraction of the cost, employing a small fraction of the people who used to support the classified ads in newspapers.

I'm not surprised, and if you think about it, it's very predictable. It's called productivity, and it's what new technology is supposed to do. We used to employ 20 percent of the workforce in agriculture, now it's just 2 percent. That's because of technology. You may say it's bad, but there's also less hunger in the US now than there was then. And there probably are far more classified ads today, now that they're mostly free, than there were when they cost money.

It's productivity. It basically a good thing. And as long as we invest in progress it's inevitable.

Here's an MP3 of the segment quoted above.

Another test post

Sun, 19 Jul 2009 17:09:32 GMT

It's amazing how much discussion these test posts get over on FriendFeed.

I have to do them because my app needs something new to ping about to see if the apps that are subscribed to this get the updates, in real-time of course, via RSS.

A test case for RSS Clouds

Sun, 19 Jul 2009 00:23:54 GMT

I'm preparing a test of a Twitter-like service based on rssCloud, but then realized that I'm within striking distance of something simpler -- re-adding a <cloud> element to Scripting News. It had one for quite a while, we had a server that supported it when Radio 8 shipped in 2002.

So I added it in, and before I could construct a test, some aggregators started registering handlers! In other words, the old network started rebooting. If you don't understand, don't worry about it, but if you do -- man that's spooky.

Anyway, the RSS feed for scripting.com does indeed now have a <cloud> element, and if you happen to have an aggregator that knows what to do with it, please feel free to register a handler. I'll have a lot more to say about this in the coming week, knock wood, Murphy-willing.

BTW, the errors in the log are the result of apps registering receivers who are unable to receive notification because they're behind a firewall or NAT. I obviously have to unsub them when I get that error.

And that's the way it is...

Sat, 18 Jul 2009 00:29:13 GMT

NPR: Walter Cronkite, The Nation's Narrator.

NY Times: Walter Cronkite, Iconic Anchorman, Dies.

This is not an earth-shaking announcement

Fri, 17 Jul 2009 22:40:33 GMT

I'm just one guy programming away, but I have met Valentino Rossi, and when I'm very productive, as I have been this week, I feel like I'm programming the way he rides a motorcycle.

When things are clicking, programming-wise, it activates other parts of my creativity. I cook more imaginatively, and I consider moving to Italy, so I can enjoy the good life while reorganizing the world. They have Internet in Italy so it's hard to imagine how Berlusconi could interfere.

Anyway, this is not an earth-shaking announcement. I just got something working today that I imagined for quite some time, and it's nice.

The goal is to have a Small Pieces Loosely Joined equivalent of Twitter.

I know now that there are people at Google who share this vision. They have the resources to do centralization. What we have to make sure is that the Rest Of Us have the ability to route around the centralization. I hope they don't take it personally at Google, but enough with letting one company control the flow of the real-time web. There are always pundits who are willing to sell us out to the BigCo's, but I am not one of them. Never have. I remember when Google was One Of Us. Hopefully that thread still runs strong inside them now that they are a BigCo.

While everyone was debating the morality of Arrington releasing the Twitter info, I was thinking "Geez these people are focusing on the wrong stuff." The real question is how big TwitterCorp plans to get while holding the control tightly within the confines of their Corp. That can't work. It never has. I'm amazed that smart people like the people who run Twitter are willing to bet on that, still, so far into it. That they think a single company can run the Pulse of the Planet is a sign that they are drinking too much of their own Kool Aid. This can't work. Can't.

Anyway, here's what I have working.

A Twitter-like RSS feed with a single subscriber who gets notified by the cloud when the feed has updated. It then reads the feed and displays the new stuff. This all happens before I can refresh the page. It's the same speed as the connection between Twitter and FriendFeed. Now there will be people who say it can't scale. 1. They don't know. 2. They might be right. And even if it's slower than Twitter, it's worth the tradeoff. Because Twitter is going to break. Be sure of it. Nothing in the history of the Internet has ever done what they're trying to do. I don't know for sure, but I suspect it can't be done. And even if it can, it's bad.

Anyway, time for me to enjoy what's left of the day. It's gorgeous in California. I'm going to get some exercise then have my Italian dinner and then more work tomorrow! Buongiorno and arrivaderci!

BadHair at 7PM

Fri, 17 Jul 2009 00:37:37 GMT

It's been an interesting couple of days with TechCrunch first teasing and then releasing internal notes from Twitter Corp meetings.

At first the debate was over the propriety of TechCrunch releasing this information. But now that at least some of the information is out there, the discussion is turning to the information and attitudes they reflect.

Were having a Bad Hair Day podcast at the normal time, 7PM Pacific this evening and we'll be talking about the release. You can catch the show live on BlogTalkRadio or as a podcast shortly after the completion of the show at 7:45PM.

Our guest for this evening's podcast is Chris Saad, who we had originally scheduled to talk about commenting and blogging and everything related, which is just about everything in community software.

There was a fair amount of TechCrunch agenda in there. I didn't get that they're at war with RSS from reading the TwitterCorp notes, just that they don't feel all RSS content should flow through Twitter. I concur. Twitter and RSS are used for different purposes, and there's far too much new stuff in RSS for Twitter's system to handle.

After reading the TC piece about Twitter I thought: "How wrong that a company owns this medium." Always felt that, reinforced it.

Wonder what Doc Searls thinks about this. Seems we're derailed from the Cluetrain.

If you have comments, please post them here.

Rebooting the RSS cloud

Tue, 14 Jul 2009 23:23:01 GMT

In the early days of RSS, we had the idea that instantaneous updates would be the next step. That was 2001. It took a little longer than we thought, but now with "realtime" as the Next Big Thing, it's time to reboot all that lovely stuff.

http://rsscloud.org/

If it works, it'll be a bootstrap. That means at first the results will be a few sites pinging each other and updating in realtime. If it gains traction, it'll get support from a lot of tech and media companies. It could happen very quickly.

One thing's for sure -- we'll need implementations in every language and runtime. I'm doing mine in the OPML Editor, of course. But just think of that as a reference implementation. I think the really scalable versions will be in Python, PHP or C.

Also thinking about using Amazon's SimpleDB to store the graph. We'll see..

Mulberry ice cream

Tue, 14 Jul 2009 07:00:00 GMT

The editorial team of InBerkeley.com had lunch today at Saul's in North Berkeley. I had a bowl of cold borscht with sour cream, and a toasted bagel with lox and cream cheese.

After lunch we all went down the block to Chez Panisse and had a Bastille Day special mulberry ice cream cone for $2.

What, if anything, did Microsoft announce?

Tue, 14 Jul 2009 03:10:20 GMT

This morning, before we recorded the RTN podcast, there was evidence that Microsoft had announced something. Apparently they had briefed the press at some undisclosed location on some date we don't know about some products we don't understand.

Last week I gave Google a ton of grief for announcing an operating system that has been shipping for 17 years and a web browser that had been shipping for about a year as a new product with the same name as the browser that had been shipping for a year.

In the podcast we talked about a river of realtime news. The analogy fits these pseudo-events in the following way. Sometime in the past weeks Microsoft held a private event, trying to build a dam on the river, hoping to blow the dam at a predetermined time earlier today, thereby creating a rush of news that would impress everyone. It didn't work because apparently the dam developed a leak in the middle of the night and the water rushed down the river of news while everyone was sleeping. No one was impressed. Sad Microsoft.

The moral of the story: Companies probably should announce products when they are new, and when ordinary people and the savvy insiders can try it all out and share their opinions. That way if the product is any good it will generate interest. If it's not good, no one need bother get excited.

ReadWriteWeb: Microsoft Office Web Access Not Here Yet.

Marc Canter leaves California

Mon, 13 Jul 2009 23:19:45 GMT

I remember when Marc came to California in 1988. He got here just in time for the Loma Prieta quake. Two people died outside his office on Townsend. They had just been to visit Marc.

There would be no tech industry South Of Market if Marc hadn't moved his small company here from Chicago in the late 80s. He was young then, he had a purpose, he was going to turn desktop computers into movie machines. He did that.

He was hugely influential, although time has a way of paving that over. Remember that, young people, you may be important now, but there will come a day when no one knows your name. Prepare for that day.

He got rich, very rich -- but he blew through the money like a drunk rock star with an entourage, which he had.

Marc is a wild man. California has captured his wildness for a long time. Now that wildness belongs to the rest of the USA.

Good luck man -- keep blogging so we know what's up with you.

Rebooting The News #16

Mon, 13 Jul 2009 23:16:43 GMT

Recorded this morning at 9AM Pacific.

Show page here, with notes written by Jay.

More Pubsubhubbub feedback

Sat, 11 Jul 2009 16:25:09 GMT

When I travel to Europe, I wonder why they couldn't just do electric plugs the same way we do in the US. That way I wouldn't have to carry an adapter and I'd be able to plug in more than one device at a time. I wish their cell phones worked the same way ours do (I gather they do now, somewhat) and that billing worked the same (I'll let you know when the bill from my June trip arrives). When I travel to London I wish they had the good sense to drive on the correct side of the road.

Each of these inconveniences were caused by engineers thinking they didn't "have to" worry about the way things were done before. They were right, they didn't have to, and all future users paid for their insistence. Think how much better it would all have worked if they cared.

And some things are, thankfully, the same. For example -- a wifi router is the same in Europe and the US. The Euro is a way of rolling up currency incompatibilities, although some countries in Europe, Denmark, the UK and Switzerland, aren't on board. But think about all the trouble they've gone to get that compatibility. What if they had been compatible from the start?

Anyway, how does this apply to notification?

Googler DeWitt Clinton asked for Feedback on Friendfeed's proposal for notification, which is different from Google's. I'm already confused! Both of them are different from the weblogs.com method which is now almost ten years old (and deployed in every blogging app and CMS out there).

I make the same suggestion to them that I made to the IETF when they were embarking on Atom. I offered that they should start with RSS 2.0 and change whatever they felt they can't live with, and document their rationales. They didn't take my advice, so now we're in this silly situation where there are two names for everything. What RSS calls an <item>, Atom calls a <froofraw> (or whatever, I can never remember).

2003: Prior art as a design method.

So, if you're working on notification, I suggest starting with weblogs.com pinging with changes.xml as your output, and then change whatever you feel you can't live with, and document your rationales. That way what you end up with will be minimally different from what's already out there, and future implementers won't curse us for not having the sense to have one way to do things. (That's right, they'll curse all of us, they won't know or care who went first.)

Now, if forced to make a choice, I'd probably go with Pubsubhubbub for three reasons: 1. It's at least XML, even if it's not RSS. 2. They say they'll support RSS, giving a sense of being in touch with the world they live in. 3. It's Google, so they have a certain amount of sway with users and developers. However, neither of them adopts the prior art method of format design outlined above. If either of them did, I wouldn't even have to make a choice.

A more low-tech approach to ping hubs

Sat, 11 Jul 2009 03:15:09 GMT

When talking with the Google guys earlier today I told them that there was an even more low-tech approach than the <cloud> element for the kind of notification they were doing. As I was reading their spec, I decided to look into it to refresh my memory. I'm writing it up here, so everyone can compare.

1. Unlike <cloud> this protocol was very widely implemented. Support for this protocol is already baked into almost all blogging software, and (likely) many CMSes.

2. The feed indicates which hub it belongs to using a <category> element. You can see an example looking in the feed for my Radio weblog. I've made a copy of that feed in case the link goes bad (I hear that Radio weblog hosting may end in December.)

This means that if you want to find out if this feed changed, you should monitor the indicated changes.xml file.

3. When the feed updates, it pings the server that maintains that changes.xml file. The coupling here is much looser than the coupling that Google is using. But the changes.xml file can be read once a minute. If your application can handle up-to-the-minute updates instead of up-to-the-second, then this approach works fine.

A neat conference hack?

Fri, 10 Jul 2009 23:00:17 GMT

Hey I just had an idea for a conference hack you can do at a traditional audience-oriented conference.

I have a theory that you could grab any random person from the audience and put them on stage and they'd give a better talk than the usual conference speaker because they wouldn't have had time to prepare slides or get nervous and plan a speech in their head.

So, why not do exactly that!

Have a "surprise panel" mid-afternoon on the first day, around 3PM. The conference moderator takes the podium and says: "Would the following people please come up on stage." And then he'd name four people chosen at random from the audience. Then they'd have a discussion about the previous panels and speeches, the topics of the day in whatever industry or profession the conference is about.

The only problem with this idea is that by 3PM most of the people would already be out in the hallway schmoozing because the speeches and panels were so boring. Not exactly sure what to do about that.