Scripting News

An open letter to Google

Fri, 20 Nov 2009 21:01:05 GMT

I had an interesting but somewhat disturbing exchange with a Google guy on Twitter today. It reveals a bunch of disconnects, that I'm going to try, in this post, to address.

1. Please take these statements at face value.

2. I am just a person, I am not in competition with Google.

3. I am a Google user. My primary email account is on GMail. I just bought a Droid, and started a Droid blog to help other people get started. I like it primarily because it connects so well with Google services.

4. I am a former Google shareholder. I made a shitload of money from my Google investment. Thank you.

5. I think Google is a big company. I think the people at Google, like most people everywhere, mean well. Like every big organization there are some who don't mean well. But I judge each individual as a person. I don't assume because a person works at Google that they are good or bad or otherwise.

6. I don't have the first clue what it's like to be inside Google, and honestly I don't care.

7. Now about PubSubHubBub. When I first looked at it I saw Atom all over it. I quickly hit the Back button.

8. There was a time when I seriously considered implementing it. But it required me to understand concepts I didn't understand and had no interest in investing in. It seemed to me that I would have to reimplement a lot of stuff I already had working. This is something big companies ask you to do a lot of.

9. One of the reasons I revitalized rssCloud was to influence Google to support RSS better in PSHB.

10. One of the clues that PSHB needs to be reconstructed is that it's so hard to describe. What's needed here is easy to explain: Instant updates for RSS. If you think RSS is a bad choice of terms, do some research. The world sees it that way. If you make that more general, you lose people. They get confused. PSHB is very very confusing to people. That hinders adoption.

11. Fostering adoption of complex technologies is something I know a lot about. I'm very good at it. You can ignore me if you want, but I usually am right about this stuff.

12. Switching gears, I like the Internet because it means I can ignore big companies and still create meaningful software.

13. I think Google doesn't like RSS. I see that in a lot of things Google does.

14. I wish Google would give up on fighting RSS. I think it's pointless. I don't think defeating or blunting or obviating RSS has anything to do with Google's business.

15. You can argue with me on any of these points, but remember #2. If you convince me I'm wrong (which is unlikely, btw, I'm no different than anyone else in that regard), you still have just convinced one person.

16. All this disclaimed, we have a common interest, I think. I don't want to pretend to speak for Google, so I don't want to try to say what that is.

Where is RSS?

Fri, 20 Nov 2009 19:33:20 GMT

I watched the morning session of TechCrunch's second realtime conference, including the half hour interview with Dick Costolo, the COO of Twitter.

Of course Mike Arrington asked him the "Is RSS Dead?" question, and thankfully Costolo didn't want to go there. It would be ingracious of him, of course, because he made $100 million with RSS.

He said RSS had been "pushed down" the stack, and it was now a protocol like SMTP or HTTP.

In a way I agree with him, but only so far.

RSS was never anything more than a protocol like SMTP or HTTP. So it hasn't gone anywhere. It's still exactly where it has been since 2002, it's part of the fabric of the Internet, and is the standard format for news distribution. We're lucky to have a standard format for that.

But...

Had Arrington asked me the question, I would have answered it differently.

RSS will form the basis for the open distributed version of Twitter.

The loosely-coupled 140-character message network.

RSS already has everything we need, including a protocol for realtime updates.

Further, any vendor of a Twitter client would, imho, be well-advised to spread out to achieve independence from the Twitter company. One way to do that, and they should all do it, is to support Facebook on an equal basis with Twitter. But that isn't enough. They should all make an investment in the open distributed way of doing what Twitter does. What that means is to offer the user the option to create a backup of their tweet stream in RSS, as a publicly-accessible feed. And once there's a base of apps doing that, they should add a feature to subscribe to those feeds.

Key point: Once they're there, they can add core features without waiting for Twitter.

Of course Arrington didn't ask me that question, and that's fine -- that's his prerogative. But there's nothing to stop me from answering it anyway!

Coolest software of the decade?

Thu, 19 Nov 2009 20:48:06 GMT

Everyone's asking questions about the decade that's coming to a close, I'd like to ask what's the coolest software you used this decade?

For me, it might be Dropbox. I keep thinking of new uses for it.

For a guy with a huge number of computers (I don't even want to count them), it's not only a lifesaver but an idea factory. I've already built utilities on it. The basis: polling a folder is incredibly low-cost. You can do a lot of it without impacting the performance of your machine. That was true in 2002 when we made Radio do upstreaming. It's even more true today.

Because Dropbox wires together folders on any machine you link into it, it's a very simple content distributor. You can have 18 computers looking for something, when one finds it, they all find out and get the thing. It could be large or small.

Like all cool things, it's fairly obvious, and has probably been done many times before. But they put it together now and it works and is trivial to set up. I keep thinking of things to use it for. All of which makes it very cool. Unless I'm missing something, it's my CSOTD.

Update: There's a thread on this topic on Ycombinator.

The new Retweet is cool!

Thu, 19 Nov 2009 18:35:23 GMT

I sort of understand why people don't like the new retweet, but I like it very much, and probably for many of the reasons they don't like it.

If you follow me on Twitter you know that a lot of my tweets are links to stories on the web. I would probably forward other people's links more if there were a way to give them credit for the link without adding all that overhead to the text. I find that once you add a bit of text to a tweet you dilute its meaning. Do it two or three times and its a confusing mess. I don't know who said what.

Worse, often the meaning of messages are reversed when they're retweeted. Not only does the person show off that they didn't understand what was said, but they propogate the mistake by sending it to all their followers.

In the new method, forwarding a link through Twitter is error-free, no noise is added because it can't, and the lineage is carried as metadata, and doesn't take up any of the 140 characters.

I applaud features that don't use up the 140 characters, and like even more features that give them back to us. I think Twitter should be encouraged to do more to pull data out of the text of a tweet and carry it as metadata, so apps can do stuff with it, and so people get to use the 140 chars to say what they have to say.

I do almost no retweeting in the old regime. But I already do a lot more now, and will do even more once everyone has the feature. Once it's been out there for a few weeks I think we'll wonder how we ever lived without it.

Journalists as ski instructors

Thu, 19 Nov 2009 18:11:21 GMT

One of the cool things about riding on a train is that you meet a lot of people.

There are Europeans who are visiting the US and have the train riding habit from home.

There are people who remember the golden age of trains and can tell you how this or that is a shadow of its former self.

And there are people who are afraid of plane travel and prefer trains to buses.

There are also people like me who had a cross-country train trip on their bucket list, and found that the fantasy was better than the reality. (Partially because this trip follows the route of I-80 and I-70, which for me is well-traveled, by car.)

When you're sitting with strangers in the dining car, conversation turns to What You Do, and part of my story is Rebooting The News. In explaining what was happening with the news system in the US, I came up with a new analogy this time, which I told in Rebooting The News #33, and thought I should repeat here.

Journalism is like skiing in the 50s or 60s. Previously it had been a sport that very few people enjoyed, and they were all very good. But now the doors were opening to amateurs, as it did with skiing. The pros are going to have to share the slopes with people who don't take the sport as seriously as they do. They're still going to be able to ski, but the rest of us are not just going to admire them for how skilled they are, we're going to do it too. They can even earn a living as ski patrol and ski instructors. Or lift operators or more mundane jobs like people who work in hotels and drive the shuttle bus. There are still jobs in skiing after the arrival of the amateurs. But the exclusivity is gone.

Is Twitter more open than News Corp?

Thu, 19 Nov 2009 17:26:51 GMT

My chin fell to the floor this morning as I read a BBC article quoting Twitter co-CEO Biz Stone advising Rupert Murdoch to be more open.

This got me to think about where Twitter is and where they're going and how similar it is to where Murdoch's newspapers are.

In a newspaper, reporters get the prime space with the big headlines, and the readers are placed in a corner, Letters to the Editor. Or represented by a "Public Editor" who does a better job of representing the editors and owners.

In Twitter there's a similar hierarchy developing, pretty rapidly.

The prime space is allocated, in a totally non-transparent way, to certain people, and the rest of us are mostly talking to ourselves, in very small numbers.

I was having coffee the other day with a former colleague at Berkman, Ethan Zuckerman, who said he would try to do something special if he had the millions of followers you get when you're on the Suggested Users List. I've seen people go that route. All of a sudden it's not good enough to be yourself, now you have to do something to take advantage of the flow you're able to generate. I wonder if that distortion, when it all shakes out, will be all that different from the feeling a reporter gets that he or she is more than a person writing from their own point of view. My guess is that it's more or less the same thing.

Stone has made a mess of something that could have been great by not being tranparent. How ironic that he advises Murdoch on something he himself so badly needs to do. Pretty typical of the way the tech industry relates to media.

Anyway, I think it's inevitable that Murdoch and many others in the media business will see the need to challenge Twitter for dominance in the realtime message distribution network. I don't see Twitter as being any more or less open than Mudoch's company. The basis for success will come elsewhere.

Rebooting Personal News

Thu, 19 Nov 2009 02:00:16 GMT

One of Jay's ideas for rebooting professional news applies equally, imho, to personal news. I wrote it up over at rebootnews.com.

Traveling with electronics

Wed, 18 Nov 2009 12:48:10 GMT

See the Droidie site for observations on the tools I carried with me on my latest trip.

I'll build the refugee camps

Tue, 17 Nov 2009 15:38:08 GMT

Tim O'Reilly is going to give a keynote at the Web 2.0 conference about the War of the Web. You should read his piece, many good points, I agree with most of it.

The tech industry sure loves its wars.

And death. This is dead that is dead, everyone is dead, but me.

Isn't that every child's fantasy -- to have all the world to himself, to be able to drive any car, eat any food, play with any toy, and not have to share with anyone?

The other day I read that the URL was dead.

Anyway one thought I'd like to share.

If there's going to be a war for the web, fine, I already know what I'll do. I'll build the refugee camps. They will be very nice. Hiltons. You can have a beautiful ocean view or a view of the battlefield.

We'll all take pictures from our balcony.

So have a nice war, techies.

Maybe it's time for personal servers?

Sun, 15 Nov 2009 15:58:40 GMT

In the early part of this decade, after the first dotcom crash, a lot of us thought that we'd all have personal servers by now.

We called them fractional horsepower servers because the issues were different. Ease of use mattered more than scalability. And communication between servers and authoring tools was also essential. Hence XML-RPC, OPML and RSS.

Instead, user generated content emerged as a business model, and many people went with the free hosting offered by startups. I never have depended on it, I've been inside too many tech companies to be willing to trust my writing with them, esp not long-term. The UGC business model only seems good for the users -- as they say if the offer appears too good to be true, it probably is. If you read the user agreement, they have no long-term obligation to host it. They probably don't even have to give you a copy of your own stuff.

People ask how I use River2 while I travel. Well, my ISP, AT&T, offers a plan where you get five static IP addresses. I'm pretty technical so I know how to set it up, and I have an old laptop in my house that runs River2. I log into it even when I'm getting on from the house, but I can check what's new from an airplane at 35000 feet, where I am right now. I've not mentioned this before, but a couple of people asked me how I do it, and I told them, and neither thought I was crazy. That's a good sign.

Not that Google Reader isn't an excellent product, it is. But it isn't what I like. It's okay, not everyone drinks the same beer or drives the same car. And with broadband becoming more popular, and computers cheaper, and old laptops lying around doing nothing, maybe for some people now's the time to start looking at having your own server running in your own house.

It'll be interesting to see what kinds of comments this post gets.

PS: There's a thread on this topic at YCombinator. Major misunderstanding, by personal server I mean one that you pay for or own, it doesn't have to be running in your house. If you pay for a server at Rackspace or EC2, that's fundamentally difference from the UGC model. That's the important difference.

Another day of train travel

Sat, 14 Nov 2009 18:39:28 GMT

Woke up in the middle of the night in Salt Lake City, went back to sleep, and by dawn we were in the middle of a whiteout with snow on the Wasatch front. Headed east from there, roughly following the path of Interstate 70, through Green River and Grand Junction. We'll get to Denver at about 7PM, which is where I will get off the train, and head to the airport tomorrow for a flight to an unnamed destination to hang with friends for a few days.

Taking pictures all through the day!

Why the collection is important

Sat, 14 Nov 2009 00:11:18 GMT

In response to my post about the new editorial tools I am using, Bill Seitz asked why it's so important to have a representation of the pre-rendered content stored in public on the web. My first answer was incomplete, I said I wanted an archive. I don't feel comfortable having the only copy of things I write reside on servers of corporations who might decide at some point they're not interested in continuing to store the stuff, or might have a technical failure and lose the stuff. Or whatever. Praise Murphy.

But there's another even more important reason. I hope that at some point we might swing back with everyone having their own home base and that we might still have the benefit of real-time updates, and scatter the bits all over creation. I want the best of both worlds. A place where all my writing is collected and preserved and can be commented on, and having that same content appear in as many other places as people want to view it. This was the point of syndication in the first place, to give people lots of options for viewing. And while not many people knew about the cloud element in RSS, it was there since 2001, so I don't think I have to work too hard to persuade anyone that real-time updates was always part of the vision of RSS. It was.

If we're going to get there, we have to start. That's what I'm doing, starting.

SF to Denver by train

Fri, 13 Nov 2009 17:59:53 GMT

I've always wanted to take a train across the United States. Today, I'm going to do a big part of it, from Emeryville CA to Denver. Not sure where I'll go from there, playing it by ear.

I don't know how much of the trip I'll document here on scripting.com, but you can see all the activity on protoblogger.com, including a set of pictures on Flickr. All part of a grand experiment to pioneer the next generation of creative writing tools for the web.

My tools: An Asus Eee PC 1005HA, standard issue (no upgrades). I'm using my Droid, tethered, and Verizon for connectivity, but have my Sprint MiFi and iPhone with me as backups. The camera is a Canon PowerShot, but I may use the cell phones for quick pictures.

I'm on the California Zephyr, have a bedroom so I'll get a good night sleep, meals included and coffee (thanks for that).

Want to know where I am at any given moment:

Working on new editorial tools

Thu, 12 Nov 2009 18:57:29 GMT

This week I set a goal to get my next generation of editorial tools to a level where I could use them for almost everything I do online. Not yet for others to use, this is how I develop stuff. I do more than eat the dog food, think of it this way -- I am the dog.

So, while I have been writing very actively online for the last few days, very little of it has been appearing here at scripting.com. Eventually I'll figure out how to migrate so that it is. Right now the place to go for it all is protoblogger.com. Which is an apt name, because I feel like what I'm doing now is the prototype for what blogging will be like in the future.

Like the first generation, the new stuff mixes linkblogging and writing of longer posts.

The first time I did this stuff, it was easier, all the content flowed to one place, a static server that I ran.

In the second gen life was more complicated, I was running a dynamic server on the back-end (Manila) and using an outliner for the front-end.

Then I went back to static on the back end, which is how Scripting News currently runs. Then I stopped linkblogging here and started on Twitter, which still must be part of my work environment, but I have a lot more to say than fits easily in 140 characters. The challenge has been to create a tool that does both, in the same place, with agility. And empowers the author. And makes it easy to scatter the writing all over god's creation and at the same time create a feeling of "home" for the author.

After Automattic adopted rssCloud I decided to look at using wordpress.com as the back-end, rather with a static server. As I explored WordPress, I realized it could solve a huge amount of the problem for me, and I had no interest in doing yet another dynamic CMS, so I embarked down this path.

I gotta say, now that it's all working, it's very fucking cool.

I have 8 different WordPress blogs and my links flow through Twitter too, all from one window. This gives me so much more power than I had before, and I suspect a lot of other people are dealing with this kind of complexity too, but I am managing it. I love the complexity instead of it being in my way. It took a lot of work, both conceptual and programming to get this right, but I'm there now.

Anyway there's no purpose to this post other than to Narrate My Work and let other developers know that this kind of editorial system is coming, and it has special needs on the back-end. There's no single rendering of the content, since it's scattered all over the place (the lifestyle of our time). But there is a new position for a static server that stores the user's full content flow. It's a low-tech workhorse of a server, but it's super-important.

I have to maintain a server for the unrendered content. That works fine for me, but won't work when I get users. So the back-ends should probably evolve to not just display the rendered HTML but to allow tools to store the source code for the writing along-side. That's how writing tools should be working, imho.

The Droidie community looks at tethering

Wed, 11 Nov 2009 19:51:07 GMT

And gets the answer.

http://droidie.com/2009/11/10/can-android-tether-today-for-0-extra/

Blogging at it's besssst.

Thanks Matt for listening...

Wed, 11 Nov 2009 14:08:05 GMT

Just got a link from my book agent Steve Hanselman, to this piece by Matt Mullenweg, on CNN -- 10 blogs to make you think.

I was really proud to see my humble blog at the top of his list.

And I'm proud of Matt -- he's done really well with WordPress. I'm using it all day every day and building my newest software around it. Why? It's pretty simple, and Matt says it in his piece. He listens.

And when I asked him to add a feature to WordPress he said yes. I didn't even have to call him, or buy him lunch -- all I did was ask on the blog. He must have been reading.

There isn't enough trust in the world, imho. People can't tell, or don't take the time to find out, if someone is trustworthy. The other day I asked this question of an editor at a major newspaper -- why don't you trust your readers? I ask this of Apple, why don't you trust your users? What about the government, why doesn't it trust its citizens? Ultimately all these institutions must listen to the people they serve. The news and tech industries, even governments -- will eventually listen.

The reason people are reluctant: If you extend trust, sometimes you're going to get burned. And if you never trust anyone, you'll never be hurt. But you won't have much of a life. So you have to develop a sense of who and what you can depend on.

Not many guys in Matt's shoes take a chance on a guy like me. But it just takes one to make some amazing things happen! And while today's news people don't seem to trust me, all it took was one to revolutionize how news flows through the Internet. Just one. That's all.

I look back to the times when I have been most effective, it's always been in partnership with someone else. That's the big secret. Take a chance, and when it works, take another, and another.

Pretty soon I'm going to put another invitation out there to the universe, and I know I'll get a listen from Matt, and I hope from some other people too.

No support on Twitter please

Wed, 11 Nov 2009 13:10:14 GMT

Support on Twitter can't possibly work, for two reasons.

1. Can you really explain the problem in 140 characters?

2. Can it can be solved in 140 characters?

Better: Find a way you can ask in a comment or email, and explain carefully what went wrong.

Two bits of movie dialog

Wed, 11 Nov 2009 12:36:24 GMT

For some reason there are two bits of movie dialog stuck in my head.

1. In Fargo, there's a scene where a random cop is talking to a bar owner who's shoveling his sidewalk, telling the story of the "funny lookin guy" played by Steve Buscemi. At the end of the story, which he just told in a beautful midwestern "you know you betcha" way, when he runs out of story, he says "That's it. (Big pause.) End of story." The moment wasn't awkward at all, quite dignified, beautifully done.

2. The big "my sister" moment in Chinatown with Faye Dunaway and Jack Nicholson. That might be the most perfect bit of dialog in all moviedom. I don't want to spoil the plot by saying why, but... Wow.

The video of the scene is on YouTube. But don't click the link if you haven't seen the movie yet.

Some movies you can only watch once, these movies never get old. For some reason Fargo works really well on mobile computers like the iPhone or Droid.

Twitter for Content

Tue, 10 Nov 2009 21:08:23 GMT

Ries & Trout wrote a series of books about Positioning.

I love these books and have written about them many times.

They explain markets in terms of metaphors that help you visualize that markets work differently from the ways we were raised to think they did. Markets are not about features, or about what you remember, they're about the map in people's minds, and about the impressions products leave, not the details.

In Marketing Warfare, they depict the marketplace as a battlefield, and used the principles outlined by the Prussian general Carl von Clauswitz in his writing about war. Minds are where the war is fought.

Then they depict the market as a collection of ladders. On each ladder there's a number one, two and three in any market. Every marketer thinks his or her product is unique and stands alone, but what's important is what the prospect thinks. In colas there's Coke, Pepsi and everyone else. Poor 7Up wasn't even on the ladder, so they invented a new one called Uncola. It worked (but it usually doesn't).

A creneau is one of these new ways of explaining something so that it stands separately in the mind of the prospect. Some creneaux exist, like laptop computers and desktop computers -- we all know the difference. Some creneaux don't exist, though marketers would have you believe they do. "The leading realtor west of the Mississippi and east of the Rockies." Yeah yeah yeah. :-)

What about ladders and creneaux in markets that are developing right now. Yammer proved that there is a segment you could think of as Twitter Behind the Firewall or Twitter for Workgroups. The product and company are doing well, because that is a real segment and they are the top guy on the ladder.

Some products are new but so useful that they pretty much form the whole market. Dropbox is an example. That means one of two things may happen: They may add a feature or find a new way to explain it that puts it either into a new segment of an existing market or on the ladder in an existing market. Ries & Trout believe they would do better if they did one of those. Either be second guy on the ladder in a booming market, or split off a piece of a market and own it. Standing alone isn't such a hot deal for the first guys in a market. Just ask Cromemco, Altair and Radio Shack about their leading positions in the personal computer market in the 1970s.

So what about Droid. It does so much, it's really hard to figure out what creneau it might be occupying, so I think it's on the iPhone ladder, maybe #2 or #3. Probably #2. Call it the Cellphone as Style Statement market. The other one is probably the Palm Pre. Microsoft, Blackberry and Nokia are on the old ladder, the one that the iPhone refused to get on.

Then there are creneaux that I'm sure are there, but with no products in them, yet.

Communicating Cameras. Oh boy what a great market that's going to be when someone goes after it seriously. No one has, yet. The iPhone is a dress rehearsal for the real product whose communication ability will be as seamless as the Amazon Kindle. The tech industry hardly notices that Amazon has solved a hard problem, in typical Amazon fashion, completely. The Kindle isn't glitzy like the iPhone or Droid, but it works so well you could say It Just Works. A high compliment.

Others: Checkbox News and Social Cameras.

Another creneau that I've been yammering about for years, which I called Payloads for Twitter, I'm now conceiving in a different way -- I'm giving up on Twitter doing this -- and instead hoping that Dropbox may get there first. They have already done an RSS feed for changes to dropboxes. And they have a public folder in every dropbox. If they do a minor cleanup of their RSS and support a realtime protocol such as rssCloud or PubSubHubBub, they will be squarely in what I think of as a new creneau with enormous potential -- Twitter for Content.

PS: This piece ran earlier today on my Unberkeley blog. There are some comments there you may want to read.

Three Droidie pieces

Mon, 09 Nov 2009 16:23:00 GMT

I wrote two pieces over the weekend on the new Droidie blog.

1. Net-net: I love the Droid.

2. The holy grail in communicating cameras.

3. Press Up to Play? Huh?? Wazzat!?