Seattle burnout: RSS and OPML, and transparency in tech

“A Seattle burnout podcast after days of heavy programming on the OPML Weblog Editor. Turned out a lot longer than I planned. Yeowza.”

Dave discusses his recent work on a blogging tool for the OPML editor, which he has just released to testers. He is excited about the progress, though there are still some bugs to work out. Dave plans to give a 3-hour presentation at an upcoming conference, covering topics like the evolution of RSS and how OPML differs from RSS as a younger “sibling” format. He emphasizes the importance of having tools that generate content in a format, and how OPML is an open-source outlining platform with the potential to attract developers. Weiner also shares his views on the importance of transparency and simplicity in technology, arguing against obfuscating XML and instead making it easy for users to understand.

Good morning everybody. This is Dave Winer and it’s, wow, what day is it? It’s the 23rd of June.
I had to think about that. 23rd of June, 2005. And here I am sitting in a hotel in Seattle.
And, wow, I just, it just topped off the blogging tool that’s part of the OPML editor and released it, documented it, released it to our testers and then put a notice about it on scripting.
All this has happened since waking up at about, well, 3. 30 a. m. , 4 o’clock maybe, somewhere in there this morning , which of course is actually kind of a reasonable time. You know, I should be using a better microphone.
Hold on a second, can we just plug this thing in? This is low , man. Yeah, let me plug it in the right place.
Okay, anyway, yeah, so all that happened and it was reasonable time because of course I’m still sort of like on East Coast time.
And now I’ve got like, what, you know, I’ve been working in this sort of a panic mode, but loving it.
I mean, this has just been a really incredibly productive period for me and a realization of a lot of things that I’ve wanted to get going.
I wanted to get a, you know, small pieces loosely joined blogging tool together, one where things just happen to show up in the right places to work with all these other pieces that are looking in the right places and reasonably error proof for users.
I think we’re getting there. I mean, it’s still, you know, it ’s the update process has created some glitches.
And I don’t know exactly where the glitches are because I’m not sitting with the people where they update to know what went wrong.
But now things are going to settle down. And, you know, my next, my next stop is there’s a cocktail party this afternoon, this evening, along with registration.
And then I guess we’re something about dinner in here.
And then tomorrow morning, I get up and we sing a song and then I’m going to talk.
I’ve got like about three hours worth of stuff to talk about and I don’t know how much of a discussion it’s going to be.
So because I’ve got a lot that I want to cover the basic areas are, you know, well, where is R SS at? How is RSS evolving? There was some stuff that Microsoft is going to talk about in Dean’s presentation right after mine that I know what they’re going to talk about.
And I think it’s good stuff. So RSS is certainly huge. I mean, obviously, a lot of work going on around RSS. And I think that it’s grown up quite well.
It’s grown up nicely, not perfectly. I mean, it’s, you’d have to say now it’s a, it’s, you know, it’s a teenager getting ready to go to college kind of and it got reasonably good grades.
It got into a little bit of trouble in, you know, as a pre adolescent, but sort of straight ened out and, you know, got into MIT, but, you know, there was a , it was kind of touch and go there.
And my hope now is that it shoots straight towards being a valedictorian, but I think that it’s going to depend on the industry as a whole taking good care of it.
And I’d say the signs are good, you know, as far as that’s concerned, they’re, the kind of struggles that the technology industry went through in the 90 s between, you know, the warfare between sun and Netscape and Microsoft, Microsoft, you know, it was actually the war between Microsoft and everyone else.
In fact, they even had an acronym, what was it, noise, N- O-I-S-E, so it was Netscape, Oracle, if the I would be Intel , S would be sun, and the E was everyone else, so it was Nets cape, Oracle, Intel, sun, and everyone else.
But now, Microsoft is a very different company today than it was then, you know, I think that the Microsoft of the mid 90s was struggling to become not the Microsoft of the mid 90 s, but to be Microsoft of the late 80s.
And I think that today’s Microsoft is not trying to be that, although it has a memory of it, of what it was like before Windows became the dominant operating system and changed their reality so substantially.
But now they’ve accepted that responsibility and are, have a whole different set of problems now. The problem is that everybody in the company thinks very long term and it’s like moving this huge ocean liner, you know, and there was an interesting moment at the meeting that I had there in April where they showed me some software and I said, you know, they told me what the ship date on the software was and I said, well, it seems like we got a lot of time to do some of that.
We got a lot of time to do some iterating, tweaking and fixing whatever, you know, great, it’s good that we have so much time.
And they looked at me like, what planet do you come from? And they basically said, no, no , no, we delivered those bits already and now we’re working on the stuff that’s going to be shipping, you know, 18 months after that.
And so my spirits dropped. I had this burning desire five days ago to put a blogging tool into the OPML editor and here we are five days later and it’s in. It’s now in users’ hands, there are blogs popping up, you know, there are bugs but the bugs will get fixed in probably relatively short order.
There are a few missing features. There’s a world of difference, a world of difference. So what you’re seeing here is a desire to not to change. I never want to get, I don’t want to be in work in their environment with their realities and they don’t want to work in my environment with my realities or they can’t or whatever.
I mean, that’s okay, but that doesn’t mean we can’t work together. And we’ve tried before, believe me, we have tried. I mean, there was soap and this is exactly where we got hung up on soap.
So the goal here has been to go for some short strokes, a little step that could be taken that could conceivably have very big impact to get a process started and there will be quite a few surprises, I think, still.
You know, I did write a sort of intro to what will be talked about tomorrow, but even with that there were some major points, major points that I didn’t talk about. So, you know , like I said, I’m going to pay close attention to what they say and I think anybody who’s interested in RSS should as well.
But at the same time, there’s O PML. And OPML and RSS have different purposes, different aims, but they’re very sort of, it’s like RSS is the older brother and OPML is the younger sister.
So they’re different. They have different flavor, you know, and they’ve had different lives. R SS has, like many first-borns, has had a rougher road. It’s had to sort of invent stuff that OPML could just sort of cruise right into.
Because we knew, if we, you know, how the story goes, if we only knew then what we know now , well, that’s, in the case of O PML, there were things that were understood that one has to do to be sort of protective of a format and to give it a good sort of beginning in life.
And I think OPML has had that.
And so, well, there, it does need, there are, you know, it’s not perfect. It’s got a few flaws here and there. It’s in reasonably good shape.
And it also has another sort of property to it that’s required for success that RSS also has, which is that they are, like R SS has a tool, which is basically blogging tools that can generate RSS.
You know, an XML format that doesn’t have a tool that generates it is an XML format without much of a future. And there have been lots of them developed that were not sort of , like, driven by the development tool.
The assumption has always been, well, there’ll be so much of this content and so much interest in it, it’ll be so successful that, of course, the tools will show up in great numbers.
But surprisingly, for a lot of them, the tools have not shown up, and nor have the numbers grown to be that large, and nor , even if the numbers were that large, would there be tools available for some of the formats.
For example, HTML has not l ended itself towards formats, excuse me, towards tools that create files in that format.
There are, I mean, of course, huge numbers of tools. Every blogging tool, of course, produces HTML.
But that isn’t HTML in all of its glory, and it’s a subset of it, and there’s no generic tool that says, “Here is a great tool for creating HTML. " Now, on the other hand, like, take a word processor or a spreadsheet, whatever file format they use, they’ve got a great tool because they started with the tool first, then came up with the format second, which yields a better sort of relationship between the tool and the format.
Because the important thing for driving the growth of a format, if that’s your goal, which really doesn’t seem like there ’s much point in having that be the case, I mean, a format in itself has no particular value other than, you know, what it enables.
And that, you know, so the formats that have tools do better than those that don’t.
And OPML has always had a tool, but the tool hasn’t been evident to many, if not most people, that would be interested in it.
It still isn’t there, except you can sort of, like, feel its existence. At some point in the next few weeks, it will become a public thing. It is an open source project.
The code is licensed under the GPL, and it will be freely downloadable and forkable.
I’m going to keep moving a fork forward, but if you don’t like what I’m doing, you’re free to start your own fork. You’re not free to make it commercial, though, because it’s GPL.
But you are free to move it and involve it whatever direction you want to.
See, I think this is interesting not just because it ’s a file format that has a tool , but it’s an open source project that is a tool.
Before we attract developers, we’re attracting users, and we ’re going to keep honing it and keep simplifying it and stream lining it and providing more valuable tools that run inside of it.
We’ve already got the blogging tool, we’ve got the instant out liner, we’ve got the directory tool. These are all things that will get easier and smoother and simpler, easier to use.
And we’ll attract more and more users. And the goal is, well, of course, that is the goal.
The goal is to create lots of content in a format that has enormous growth potential.
And as a byproduct of that, to attract developers so that the tool can get better and better and better and easier to use and more powerful and higher performance and do new things, be innovative.
When you add all this up, what is it? It’s a platform. And it ’s a platform for applications that share one thing in common that they all build on out lining.
It’s an outlining application platform. And actually, if you think about it, it’s actually there’s a rich tradition here that it’s following in.
Text tools have very often been either the underpinnings of a platform or the keystone of a platform, sort of the initial app.
If you look at Notepad or Simple Text, Notepad on Windows Simple Text, they called it Teach Text actually in the early days of the Macintosh.
Those were two Windows and Mac intosh, absolutely essential basic editors, nothing very ambitious about them, which is great, but incredibly useful.
It was the fallback. You always knew if you had a text file, you could open up with Notepad.
Notepad was your guarantee that there was no lock-in. If you could edit it in Notepad, you could edit it in anything.
On Unix, they had Emacs and a whole host of other similar editors that all had programming languages in them.
And you could build applications off of them, and programmers certainly did do that.
And this is another one of those. It’s a text platform. It ’s very, very simple.
It’s file format. It’s brain-de ad simple. And creating another tool that reads and writes OPML is also brain-dead simple.
And that means that every tool that you develop, it doesn’t have to be this way, but it’s true of every tool that we have developed.
And any tool that is developed that isn’t this way, we’re going to look at it really carefully, is that the parts are easily replaced.
It’s small pieces loosely joined. That was a great line.
I think David Weinberger coined the expression.
He has a book by that title.
This is sort of how you guarantee as a user that you’re going to have choice, because if the pieces are small and the joining is loose, that means the joining is simple. It’s easy to understand . It’s easy to replace.
Which goes to I was trying to explain the other day to people why I don’t like style sheets that obfuscate the RSS.
I know I’m really changing subjects. This is kind of how I do and, of course, this is going to be my talk, although I ’m talking straight from my notes for the talk.
Although I’m leaving out huge numbers of things, and I’ll pick those up and do those in the talk.
Maybe I’ll repeat it. Maybe I ’ll forget some of these things.
Maybe it won’t be at the talk.
Maybe the webcast won’t work.
So I was trying to explain what it is that makes me uncomfortable about style sheets that obfuscate the XML.
Because the story goes like this, is that users click on these white on orange XML icons , and they see a bunch of gobb ledygook, and they don’t know what it is, and they go crazy.
I can’t handle it. Technical.
Keep it away from me.
That’s one school of thought, and it’s a very radically different approach from the school that I belong to, which says work on the format and make it simple.
Make it so that when a user sees it, and if they look at it a little bit carefully than other than just sort of glance at it, then they go, oh, okay, I get it.
That makes sense. Okay, here’s a thing that’s called Webmaster . Here’s a thing called Managing Editor.
Here’s a thing called Title, Link, Description. All right, so there’s a Gwinn in there, okay? Sumi. You know, the point of R SS was to design a format that a user could look at and not go crazy.
And I’d love to do some user testing and find out actually what happens, what that crazy moment feels like to a user, and why we have to cater, why we have to design our software to the user who freaks out when we are basically disempowering the user who is curious.
I mean, what’s so hard about hitting the back button? I see, see, and then by being open, and being open means showing the guts.
You know, if I want to lift the hood of my car, which I bought a new car about a month ago, and I have to admit, I haven’t opened the lid yet.
I haven’t looked inside. But if I want to, I know that what I’m going to see inside of there are a lot of wires and tubes and, you know, stuff.
But I kind of like the idea that it’s up to me to decide whether I want to look in there or not.
I wouldn’t like it if the person who designed the car decided to seal all that stuff off for me so that the only person who could see it would be some magician who knew the magic incantation.
Because I would feel like I would be suspicious of that. I would think they’re ripping me off somehow. I want to know what’s in there. Maybe I can understand it.
Maybe they’re charging me too much money for this specialized knowledge that really isn’t all that hard to understand.
But the reason why they get away with it is because they’ve made it so that I can’t understand it, or they sort of conned me into believing that I can’t understand it.
I would prefer, I would trust technology more that says, “Hey , if you want to see how it works, we’re going to make it really easy for you to see that . " And maybe you’ll never click on the white on orange XML button.
Maybe once you see it and click on it and go, “Okay, well that ’s not for me. I’m not interested. " Maybe then you learn, “I don’t need to click on that. " People aren’t stupid. And people who design software for people who are stupid get what they deserve, which is stupid users.
I prefer to design software for people who are smart, because I ’d much prefer to work with people who are smart.
And sometimes people who are smart act stupid because you treat them like they’re stupid.
So I prefer to treat them like they’re smart. I think you’re smart enough to know that if you don’t like looking at the XML, then you’re not going to click on the damn button.
I’m not going to worry about you. And if you do, I’m also going to trust that you know how the back button works.
And basically, if you’re using the web and you don’t know yet how the back button works, well , then you better learn how the back button works, because you aren’t going to be getting very far without it.
You’re going to end up in some blind alleys and, you know, we all did that in the first day we used the web, and then we learned that the back button was the way you get out.
As long as it works, I think we ’re fine. So I am proudly on the side of people who say, “Don’t hide the XML. Show it to them. " But at the same time, work your ass off to make that XML transparent and simple, and don ’t hide any arcane information that people have to understand in order to even be able to use the information.
Make the required information absolutely brain dead simple.
And if you ever want to argue with me that a format that I’ve designed uses terminology that ’s too complicated and you have a way to make it simpler, it’s not enough just to say it ’s too complicated because, okay , great, maybe it is too complicated, but I can only process that in a productive way if I also know what’s a less complicated way that also works.
Because sometimes something may appear to be, it’s like Walt Mossberg said of a pair of headphones. I don’t usually read the Wall Street Journal largely because they say irritating things like this.
Walt reviewed a set of headphones which were made by Logitech. They were a placement for headphones that people use for iPods.
And he said, “The problem with these things is that they’re too expensive. " And he just didn’t explain why they were too expensive. Too expensive for what? In other words, there are headphones that are cheaper, that do the same thing, that are better. No, he didn’t say, they’re just too expensive.
Oh, great. So if you tell me this is too complex, then I’m going to want to know in relationship to what? In other words, maybe it’s exactly as complicated as it needs to be in order to get the job done.
We kind of got intimidated by that kind of thinking. There was a time when I would have gone, you know, maybe they understand something that I don ’t know.
But I’m too old for that now. I think I’ve seen everything. So you need to tell me what it is, not just that something’s too complicated, but you need to tell me how to make it less complicated.
And I don’t think that the obf uscation - you see, I couldn’t go into this great detail in my keynote - I don’t think the obf uscation of the XML that, you know, like Feedburner does, or you’ll see another guy that does it soon.
You know, it’s not like the end of the world, but it’s really irritating. I don’t think what they’re showing you is a solution, because I don’t think it’s simpler.
I think that looking at that, a smart person who is not technical would go, “Well, but this looks just like the page I came from. What’s the difference?” They try to explain it to you in words, but somehow the words don’t come through. Why not just let the goddamn symbols speak for themselves? Okay, when you click on the white on orange XML button and it doesn’t take you to something that’s obfuscated, there’s no question that there ’s a difference between A and B, that you came from some place that you understand, you went to some place new and strange, isn’t the end of the world.
Anyway, one of the things I’m going to say, I’ll try to remember at the beginning of my talk to try to head off the kind of problem that I ran into in Nashville, is I’m going to say, “Hey, I have opinions. I need to disclose that up front.
You don’t have to agree with them, but in fact, I would be surprised if you agreed with all of them, because some of them, even I would admit, there are arguments on the other side that are good, and sometimes I can’t even see the arguments on the other side that are good, but then somebody shows them to me and I go, “Hmm, those are interesting arguments. I haven ’t thought of that. " That’s what I think a reasonable person does.
So, I can accept that you may have a different point of view, a different opinion than mine, and I insist that you accept that I have a right to have an opinion that’s different than yours as well.
It’s not an insult, it’s not rude. In fact, any discussion that we have where everybody agrees on everything is just simply going to be a boring, pointless, useless discussion, because you won’t actually learn anything.
You never have a chance at changing your mind if you never hear somebody who disagrees with you, and there’s fun in changing your mind.
There’s fun in hearing new points of view, of hearing that there are different ways of parsing things, and you can respect and enjoy somebody who looks at the same situation and comes to a different conclusion .
So, people in Nashville seem to be pretty offended by the idea that I had opinions, and they weren’t necessarily their opinions, but more they were offended that I had them.
I think that we can all have our own opinions and we can still be polite and respectful of each other, and we can still have fun with each other, which means that if you want to say something bad about somebody, say it about yourself , and self-deprecating humor is always funny.
Maybe if it’s not humor, then I would disagree. If a guy hits himself in the head and says, " I’m so stupid!” I put my arm around him, give him a hug, and say, “No, no, you’re not stupid. " But if somebody is self-deprec ating humor is cool, but humor at other people’s expenses, that’s not right.
Anyway, this started out to be like a one-minute podcast here.
What I was going to do was just do a little test thing so I could test out the ability to handle enclosures in the OPML blog editor.
And of course, I can still do that with this, but now it’s going to be a little bit longer .
Anyway, I’m also going to try to remember to record my talk tomorrow, and so even if the webcast doesn’t work, you’ll still be able to hear it.
I’m going to have my microphone with me, and you never know who I might end up doing a podcast with.
So, we’re here at Nome Dex, and we’re going to have a great time, and I guess that’s it for right now, everybody, so we’ll talk to you again real soon, okay? Bye! [BLANK_AUDIO]