Scripting News

London and Paris in December

Tue, 20 Nov 2007 02:44:01 GMT

I'll be in London on December 7 and Paris on December 10.

Scoble will be in both places. The Americans come to town!

Looking forward...

"cheesecake"

Creating a maintainable and thriving web

Mon, 19 Nov 2007 17:00:09 GMT

I knew the day would come when users would wake up and realize that centralizing stuff is not good for the Internet. Today two signs that things are sorting themselves out.

1. Steve Rubel writes about the danger of routing all our URLs through TinyUrl. I love what URL-shorteners do, it's especially important in Twitter when you're limited to 140 characters to express an idea. If you have to include a link, that could use up a lot of the space you have. The problem is if everyone uses TinyUrl, as Twitter does, what happens when TinyUrl goes down or is sold to someone we don't like, or disappears forever? I admit I don't know the owners of TinyUrl and what their motives are. Their service is reasonably long-lived, reliable and quick. Even so I've written my own URL-shortener and am running it on one of my servers, and I try to use it whenever possible. However, like all my sites, this one will likely disappear within a few days of my passing. I have to maintain my servers to keep them running. A better solution is surely needed. Rubel's epiphany just exposes the tiniest sliver of the huge problem below, creating a sustainable web. We're nowhere as far as that's concerned.

2. Fred Wilson writes about how TechMeme is causing the blogs he loves to focus on the same topics. I've noticed the same thing. Steve Levy writes an article that appears in Newsweek about new hardware from Amazon, and it's an instant coral reef, within an hour or two it's the top item on TechMeme and there's a whole ecosystem of thought about it, published by people who have no information other than what they read in Levy's article. Did anything real happen here? Not very much, it's like the rush of information that appeared about Leopard in the first few days of its release. The real news becomes apparent in weeks and months, not days.

This way of doing news is a remnant, it's anachronistic, a relic of the way news used to work, when guys like Bezos and Jobs would go on a press tour, seed Pogue, Markoff, Levy and Mossberg, they would write their pieces and the rest of us would settle for the very limited and highly spun information they provided. It's not that way anymore. I'll probably write about the Amazon device, I'll probably have to buy one, and like a lot of the hardware I try out, it'll go into a box I keep in the den with other stuff that I learned a little from but never found a use for. We'll get to the bottom of it, and it probably won't appear on TechMeme. Nothing unusual about that -- in the past my blog posts didn't appear in MSM, and that's what TechMeme has become part of, MSM.

Don't kid yourself (and Wilson doesn't) the pubs that used to be blogs, Mike Arrington, Rafat Ali, Om Malik, etc are now pubs that compete with the other top entries on the TechMeme Leaderboard, and they function much in the same way. Are you interested in understanding Disqus? You'll get one brief piece in TechCrunch on their launch day, but if you find a blogger who uses it, you can really understand how it works, because they will know, and because the publishing tools are now distributed and free, you'll find out what they think. That's what's changed. The press still reflects what the press cares about, competing with other press. But the blogs, who aren't trying to climb the top 100 lists, are doing something else. We're just trying to share information with each other so we can learn, so we can use stuff better, make better choices, improve the products, and eventually create new products.

You can see this philosophy reflected in exciting new products from companies like Chumby and Bug Labs. Create open platforms with widely available development tools and let the blogs take over. Google came close with Android, and there's still plenty of time, but they don't really trust blogs at Google, like most big tech companies they trust other big companies first.

That's the revolution I've been writing about since I started blogging -- when product designs come from the experience of the people, of bloggers. It's already happened, it's so recursive you may not see it. We designed blogging itself on the early blogs. And RSS? It was a product of blogging too. Every company that Fred Wilson touches is affected by blogging, every pub that Rex Hammock works on is. Every political candidate that benefits from the vetting of ideas in the blogosphere is touched by this power. It's the old decentralization thing that the Internet does so well. The reason TechMeme is doomed to be part of MSM is that it goes the other way, it centralizes. It's almost mathematics.

How to mount a network volume on Leopard

Mon, 19 Nov 2007 01:40:41 GMT

Good news -- you can do everything you could do with a network drive on Leopard that you could do on previous versions of the Mac OS, and probably more, and it's probably faster. I'll have to let you know after a bit of time using it.

Here's how you do it.

1. Locate the network drive you want to work with, starting in the SHARED sidebar section, clicking as needed to make it visible. Or alternatively, you can use the Connect to Server command in the Finder's Go menu. You can even mount servers over the Internet this way. (This part can be very slow, but you only have to do it once.) Here's a screen shot showing the available disks on a computer named Illium. Note that two of these disks are actually folders, Leopard doesn't care. That's coool.

2. Now if you look in the upper-left corner of that screen shot, you'll see that I have chosen the second of four possible ways to view the contents of Illium. It turns out the dragging procedure people were trying to explain to me only works when you're in the three-pane view, as shown in another screen, not when you're in the other views. That's totally unintuitive, it's just a matter of luck which of the modes you happen to prefer. For people who use the second mode (like me), the Finder doesn't network very well. Click on the third of the three icons and things start working very nicely.

3. Suppose I want to work with the drive named WYOMING. Now, all I have to is drag it into the Devices section of the sidebar, and it stays there. And it's fast, no long delays. People tell me, but I haven't yet verified, that this mounting will survive a restart. If so, that's a great improvement over Tiger. Not sure how well this might work on a laptop that disconnects and reconnects to the LAN much more frequently than a hard-wired desktop computer. Earlier versions of the Mac OS tend to hang for long periods when you accidentally leave a network disk mounted on a laptop after disconnecting from the LAN.

In summary -- the three-pane view is the magic view for networking. Once you have the drives mounted, you can switch to any of the views you like, they seem to work fine.

This highlights something missing in the Mac community that we may need to provide for ourselves. When a new version of the OS comes out, we need good user-oriented change notes that explain what you need to do to get your newly updated systems to work like the systems they're replacing. I don't know how many hundreds of thousands of people have made this transition, but I can't imagine that I'm the only one who hadn't figured out how to get Leopard to mount network drives. That's pretty basic stuff, networking on Macs is something I just take for granted, I don't expect it to be mysterious.

We don't even have good language to explain how to use a Mac. A number of people just said "Command-K" but that didn't solve the problem. I knew how to access network drives, but they weren't mounting. Are these well-understood terms? Apparently not. How many times did people tell me to drag a folder over to the sidebar without making sure I was already in the right view? Did they know that you had to be in that view? For people who prefer the three-pane view, it would likely never occur to them that this wouldn't work in the other views. Why would they think it wouldn't? This is very far from "it just works."

And what are they thinking at Apple, in enabling this feature in one view and not in others? This is a basic software design concept -- The Principle of Least Surprise. If you find that users are likely to do something one way, instead of teaching them another, make the software "just work." We should strive to accomodate the user, not thwart or confuse.

An ad on Facebook

Sun, 18 Nov 2007 16:10:09 GMT

My Facebook news feed has an ad for Blockbuster.

It's the first time I noticed an ad in this location.

Apple's brand promise, and how blogging can fix it

Sat, 17 Nov 2007 20:27:45 GMT

Scoble has a piece today on Apple's brand promise that nails it precisely, never seen him hit the mark so well. Congrats. The other day at lunch I was telling the Uncov guys that despite what they may think, Scoble really is brilliant. Read this piece, I feel completely vindicated (though sometimes I read his stuff and shake my head in disbelief at how he could be so wrong).

Here's the key point in Scoble's piece: Apple makes a promise with its brand and doesn't come close to keeping it.

The promise of Apple is that everything "just works."

It's true that the Mac does work better than Windows, usually, but that can be masked by the expected breakage in a user's first transition to Mac. You expect stuff to break when you switch from Windows to Mac. You expect things to work differently. But it all comes home when you "upgrade" to a new version of the Mac OS and find that the engineers at Apple don't listen to designers, or understand users any more than the geeks at Microsoft do. The problem isn't with Microsoft or Apple as a culture, the problem is with the tech industry.

Google has it too. They will break us, I'm sure of it. If I told you how, they'd unleash a storm of hate at me very much like what you get when you criticize Apple. Even Microsoft used to have its anonymous assholes on the net who would make you feel pain for questioning their competence or integrity. Hey when they cut off Netscape's air supply, they cut off a lot of users and small developers too. Sun did it with the Java wars (Microsoft again), and Apple -- well read my piece on networking in Leopard, which may look like it was coordinated with Scoble's. It wasn't; we've both been stewing in the same broth -- the hypocrisy of Apple's marketing, the lack of humility that guarantees that everything we care about, as users, will eventually break if we trust the tech industry to take care of our needs.

The only way this is going to change, and the signs are good, is if the users take over from the press at telling the truth about these products. The people at Scoble's dinner should come out of the shadows and tell their stories publicly, so everyone else who has a problem doesn't feel that the problem is their fault.

You know, when I published my piece this morning, it took ten minutes for the first post to appear that blamed me for the problem with Leopard's networking. But not much later, someone sent a pointer to a piece by Glenn Fleishman, where he says that Leopard's networking is an improvement over Tiger's. I trust Glenn, and believe him. I just didn't know when I switched to the Mac that there were so many problems. I had to discover them myself. And many more were uncovered in the switch to Leopard. (So much for "just works.")

There is something special about Apple, but it really isn't all that present in the Mac OS. The error messages say something isn't operational, which isn't really a word in the English language (why not say it doesn't work). When I followed Glenn's instructions and enabled file sharing through the Prefs system, all of a sudden my MacBook which is running Tiger can't access the file server. I'm sure there's some reason for this that most Mac gearheads know, but they're missing the big picture -- the Mac makes a promise, as Scoble points out, that you don't have to be a gearhead to use a Mac. It's a big lie, you gotta assume the marketing people at Apple know it's a lie, and they keep getting away with it, and there's no reason for them to make it better, as long as they do get away with it.

I started blogging because people lied about the Mac, then they lied about the Constitution, then they lied about everything else I cared about. And since then blogging has taken off, so we have the tools to fix the problem, and if we wait for Silicon Valley to do it, we'll wait forever. The solution is simple -- tell the truth. Once you do, then someone else will feel they can do it too. And pretty soon the companies are going to have to do it, because as soon as they lie, there we will be to set them straight. Think of how much better our government would work if we applied this same principle to governance and then you'll understand why blogging is so important.

PS: Apple ought to feel they have an option to either: 1. Live up to the promise that their products "just work" or 2. Stop making the promise. I hope they choose option 1. And ideally they'd stop making the promise too, because there's always Murphy's Law to keep you on your toes.

Yet another beautiful Berkeley street pic

Sun, 18 Nov 2007 06:07:21 GMT

How do you mount a network volume in Leopard?

Sat, 17 Nov 2007 17:43:17 GMT

I rushed through this in my piece about Leopard a couple of weeks ago, I do things other than review software, so I don't always have enough time to go into depth. And I wanted to be reasonably sure it was as bad as I thought it was. But now I am reasonably sure, but maybe I'm still missing something, if so, I bet a lot of other people are too. Here's the problem with networking under Leopard.

In the previous version of Mac OS X, you would mount a remote volume, and from then on it was as if it were one of your local disks. That's how networking has worked on Macs since the 80s, and it's the way it works on Windows (not sure when it came in there, but it was present on NT and XP). It's the way networked OSes should work, it's hard to imagine them not working this way.

However, amazingly, that's not how it works on Leopard.

Here's an example. I have three computers on my LAN that I can access from the laptop I'm writing this piece on, Bucharest, Darkstar and Illium. They are conveniently listed in the Shared section in every Finder window. This is a small improvement, in previous Macs, you had to 2click on a Network item in the same place, and choose the computer from a dialog. Now you can see the names without clicking (It's a small improvement because believe me, I've got these names memorized.)

Let's say I want to look at the disk named Ohio on the computer named Darkstar. I click on Darkstar, and a list of disks appears, among them Ohio. I double-click on Ohio and the list of disks is replaced by the files and folders in Ohio. Nothing has changed in the left pane of the window, no disk has been mounted, I can access the contents of this disk only in this window, and only as long as it stays open. If I navigate to another disk or folder, I no longer have access to this disk. This is the first major step back. (There were some minor reverses on the way here, each of the steps in this process take much longer for some reason than they did on the earlier version of the OS. I have two machines that haven't been Leopardized, so I can compare, and the delays can be really long, and yes, I've rebooted everything numerous times. The pre-Leopard machines are faster. I actually replaced one of my Mac Minis because it was too slow, now after "upgrading" it's just as slow as the one it replaced. Oy.)

But here's the real kicker. Suppose I want to save a file to the Ohio disk from inside one of my apps. There's no way to do it!

This is the part I can't believe.

I can't even go through the navigation process to locate the disk (a lot of extra steps from the old method, where I could just access it as if it were a local disk). It's not that it's hard to do, it's that you can't do it. This is a basic feature that goes back to the 80s. How do they get away with removing it, and no one calls them on it, and they don't explain it anywhere? (Or did they and I missed it. In a Steve Jobs keynote, did he say "Oh and one more thing, we removed a feature so basic you don't even realize it's there.")

Now, as I said earlier, it's possible it is there, staring me in the face, and I just can't see it. I've been using computers long enough to know that that sometimes happens. If so, show me how to do it. How do I save a file to a server volume from inside an app?

Update: You can navigate to shared disks in some apps, and not in others, as has been pointed out in the comments. Note that in earlier versions of the OS you could save to network disks in all apps.

Here's a video that illustrates how the Finder doesn't let me mount a network drive in Leopard.

My Chumby is here

Fri, 16 Nov 2007 18:02:21 GMT

I've activated my black leather Chumby named Robusto, and as I write this, it's downloading a software update through wifi.

First impression: This is a breakthough device, kind of like the Cobalt Qube was in the mid-late 90s.

1. They use BitTorrent to distribute updates. There was 1 seed and 1 peer when it downloaded my updates. Very good use of BitTorrent, and it's smart to build it in there from the beginning.

2. Whoever did the animation was doing acid in the 60s, disco in the 70s, coke in the 80s and a dotcom startup in the 90s. It's really outrageous, really good and cool and funny. It makes you laugh out loud. Can't say the iPhone made me do that. The Qube, while it wasn't trippy, did too. (Update: Susan Kare did the design. No wonder it's so great!)

3. They have a built-in RSS reader, not sure how it works, but I configured it to display Scripting News.

4. I changed the clock to a cuckoo clock (from a plain blue analog clock). The device checks back with the website periodically to find out if it's been reconfigured, and now without me doing anything else, the clock has changed. If I tap on the door the bird comes out!

5. They have a Flickr widget that is very webservice-ish, and clever and complicated. I tried to configure it to show my friends' pictures, but the authentication window never appears in Firefox. Great idea and I can't wait to try it when it works! (It's similiar to something I'm doing with a Mac Mini as a settop box.) Update: I needed to tell Firefox that it was OK for the Chumby website to pop up a window, and when I did, the authentication worked.

6. It would be nice to have a USGS earthquake widget. It could be two-way since the Chumby has motion detectors.

7. JY Stervinou sends a pointer to a page that shows you how to turn the Chumby into a web server. That's what I'm talkin about!!

8. What's playing on my Chumby right now.

9. A comment from Steve Tomlin, CEO of Chumby.

10. Phil Torrone, via email: "One thing that i have been trying to tell folks about this device is that it is a great example of open source hardware, Chumby has released the schematics and files needed to improve. I think many people will use the Chumby as a low cost Linux computer for all sorts of amazing projects."

Summary: I got it set up and running my widgets within an hour and it was fun! I love this device, it just reeks of potential. And they did a beautiful production job. It's easily as innovative as the iPhone, but it isn't getting as much attention. Take a look you won't be disappointed.

My LAN is back on the net

Fri, 16 Nov 2007 23:27:46 GMT

I've been limping along here the last couple of days running on a shared EVDO connection, which is nowhere near as much bandwidth as I'm accustomed to. The problem was my DSL modem, which had burned out and needed to be replaced.

One of the side-effects of the outage was that the nytimesriver site stopped updating, as Jim Goodman noticed. Sorry for the outage, but I'm glad to see it was missed. My next task is to see what if anything I need to do to bring it back online.

Is Twitter down?

Fri, 16 Nov 2007 00:51:10 GMT

It's been flaky all day, and they removed a really key feature, and now (4:55PM Pacific) I can't get through at all.

If you can shed any light on this, please post a comment here.

The feature that's gone is they would hot-up names that appeared after at-signs, so you could click on the name and go the person's page. Let's hope it's just a bit of breakage and not something they did on purpose.

Update: It's pretty clear that Twitter has been down for a couple of hours (as of 6:10PM). It's also clear that it's become an integral part of communication for some of us, I'm one of those people. This outage is giving me ideas for a low-tech decentralized way to do Twitter with RSS and software running on the desktop.

Update: It's back up at 6:45PM Pacific. Jack Dorsey says the @ problem is a bug and will be fixed.

Good morning!

Thu, 15 Nov 2007 18:24:22 GMT

Yesterday at 6PM my home LAN went off the Internet. The DSL service was down, first time in a year, and in that year I had built a fair amount of stuff on the assumption that the connection is there. I couldn't update Scripting News, for example, because the CMS was running on an old laptop in the den.

It took some effort but I think I have the app moved, and I'm using the EVDO card while AT&T gets around to fixing the problem, which may happen as early as tomorrow afternoon, knock wood, praise Murphy etc etc.

Just as this is happening a ton of other things are demanding my attention. It's at times like this that I wonder how I ever used to get so much done. Kind of a miracle. Life is slower these days, that's for sure. At least for me.

The net went down just as I was about to put up a picture of Sponge Bob next to the bit about startups being sponges.

Startups must be sponges

Wed, 14 Nov 2007 22:25:59 GMT

I posted this on Twitter, it's worth posting here too.

You'll never make your product better if you shut out all criticism. You have to iterate to hit the sweet spot.

Startups must be sponges.

I don't often say people are wrong, I'd rather say products are flawed, or companies make mistakes. But companies that try to shut down critics, with personal attacks, are wrong. Startups that do it are worse than wrong, they're doomed.

Watching Jay Rosen

Wed, 14 Nov 2007 19:07:07 GMT

One of the biggest scores of BloggerCon I in October 2003 was connecting with Jay Rosen, journalism prof at NYU. He predicted almost everything we're doing today with blogging, long before there was a world-wide web. He understood that eventually publishing tools would become easier and cheaper, as would distribution, and that eventually the ability to write and publish news would become more commonplace. He used different words for what we do, but we understood what he was saying anyway. He taught us so much about the value of journalism, things we understood intuitively, he gave us words for.

At Jeff Jarvis's journalism conference at CUNY in October, Jay talked about his latest idea, combining the tools of social networking and journalism, combining the expertise of the readers and community members with the resources of professional news organizations.

The project he described then is now launching as part of NewAssignment.net. I've learned to pay attention to Jay Rosen, and recommend the same to anyone interested in the future of blogging, publishing and news.

NewTeeVee conference in SF

Wed, 14 Nov 2007 18:51:57 GMT

Today I'm at Om Malik's NewTeeVee conference in San Francisco. I came in at about 10:30AM, at the beginning of a panel discussion about advertising in Internet television. I seem to always walk in for advertising panels, it happened in the journalism conference in NY and at the Web 2.0 conference in SF, both last month.

Is it just me, or do they talk a lot about ads. As you know, this isn't something I'm very interested in. I'm more interested in, as Andrew Baron says so well, in how we can make stuff that's more interesting and useful. Not into the Long Tail stuff, or maximizing sell-through and all that other very good stuff that marketers are focused on. I'm more interested in communicating with the people I choose to communicate with.

"Maximize the parameters you want to maximize."

"The most effective units are the most disruptive."

Advertising panel movie

Wed, 14 Nov 2007 19:51:42 GMT

Performance test

Wed, 14 Nov 2007 20:22:46 GMT

Incredibly high free wifi bandwidth here at SF State.

The scene at NewTeeVee Live

Wed, 14 Nov 2007 19:43:25 GMT

Still fumbling around with Flickr

Tue, 13 Nov 2007 18:48:10 GMT

After spending most of yesterday trying to understand what was going wrong with the Flickr API, I decided to look in a different direction and tested the uploading code, and it didn't work either, so I posted a note on the yws-flickr mail list, hoping that someone else would have an idea how to proceed. A few minutes later John Watson posted a suggestion. I haven't tried his workaround yet, but I will, probably after lunch.

Don Park is having the same problem with his Flickr app.

Finally, we (seem to have) solved the problem. Thanks to Don and John for their help. Here's a screen shot of the routine that implements the technique we're all using. Basically it now takes more time for the data to move around on the Flickr side. Is this breakage? Hard to say -- it is communication software, so you have to be prepared for latency and packets getting dropped. It did used to work. So it's both breakage and more robustness was required on our side. It just shows how sensitive these systems can be, and of course that they used a fancier method for security, that is safer, but also more fragile. There are always tradeoffs, that's one of the constants of computer system design.

Berkeley hills path on a sunny November afternoon

Wed, 14 Nov 2007 04:00:39 GMT

More breakage in the Flickr API?

Tue, 13 Nov 2007 03:43:19 GMT

A few days ago I reported here that some code that built on the Flickr API that had worked properly for months had broken. It turned out that the people at Yahoo had fixed a bug in the way the API worked. They were permitting anyone to download all sizes of a picture, when (they believed) this should only be available to the creator of the picture. My code didn't try to sign the user in (there was no need to), so all of a sudden I was getting mysterious errors from my code saying that it couldn't locate an object called "Original." It took some time to trace it down, and by searching on the net for other people who were having the same problem, and figure out how it related to my code.

When I got to the bottom of the fix, figured out what they changed, and remembered how my code worked, it was easy to fix. But these were hours that I should have spent fixing other bugs, or creating new features for my users.

I put it behind me until it happened again, today. Some code that had worked for a long time is now broken. I spent most of today trying to understand, again, what a token is and a frob and how they relate. I have to admit, that when I first implemented this code I didn't really understand what they these things are, but I fumbled around until the code worked and moved on. But now I'm back to where I was, and wondering whether there's any point in trying to fix this problem. How long before something else breaks?

Right now I have a very small number of users, and most of them are not affected by these breaks. But what happens when there are more users, or something changes that breaks more of them. They're not going to be so understanding, I'm not going to be able to pass the buck. I'm going to be wrong, if that should happen, for choosing to build on Flickr. Is this really a position that Flickr wants to put us in??

I'm familiar with the thinking that one should fix problems in the code behind an API, that when you discover a bug it's just like a bug in normal software. The first time I made a change in Frontier that broke developers (including myself, btw) I understood why you have to live with the bugs once people have built on your API. To this day there are bugs in Frontier, lovingly preserved. If they were fixed, it would cause an unknown and therefore unacceptable amount of breakage.

I build on top of a lot of web apps, not just Flickr, and so far all has been good, until this round of breakage. It's a warning to everyone to live with your bugs. If you really must fix them, come up with new entry-points that work the way you think they should, or employ optional parameters. No matter what, breakage is not acceptable, not like this.

And also, if you're going to be in the business of breaking developers, get very very good at communicating, and explaining carefully what the change is, that way when we're down, we have a chance of picking up the pieces quickly. I don't have any idea where to go to see a log of changes made behind the Flickr API. Not saying there isn't such a place, it just needs to be more obvious where it is.

Here's what I know about today's problem. When I call flickr.auth.getToken with a frob returned by flickr.auth.getFrob, I get error code 108: Invalid frob. "The specified frob does not exist or has already been used."

I don't know where to go from here. I'm just sending back to Flickr something they sent me. As I said above, this has been working for months. Of course I've tried searching for others who have had the problem, but I still don't know where to go. Any help would be appreciated, of course.