Feb 2004, 15 entries

Disclaimer: This will be a really dorky entry to anyone that doesn’t get excited when they hear the words “distributed identity,” “movable type plugin,” and “authentication service” in the same sentence.

What is the talkr.net proposal?

talkr will be a distributed identity system that ties Movable Type to Flickr’s authentication service via an MT plugin, and allows people to comment on talkr-enabled blogs through their Flickr account (see, I wasn’t joking). This will allow you to maintain your identity in one place, while also enabling a couple much dreamed-about features such as:

• Get notified of new comments on posts that you’ve commented on
• Watch what your friends are talking about on other sites
• PGP sign your comments without tons of hassle

There are three main components to talkr’s service:

• A Movable Type plugin that manages comments on a particular blog
• A talkr.net website that is notified about new comments on blogs that have the MT plugin and handles some preferences
• Flickr’s authentication service which allows us to organize accounts around a distributed identity

Why should we build it?

We should build it because we can build it.

How would it work?

1. Joe Blogger installs the MT plugin on their blog (it will consist of a plugin, a perl module, a cgi script, and a few changes to their comment template).
2. When someone is posting a comment on their blog, they will have the option of “authenticating with talkr.net” with the click of a button. This button will be integrated with the comment form.
3. The button spawns a popup to mt-talkr.cgi (on their server) which checks for a mt-talkr cookie.
4. If mt-talkr.cgi finds a cookie, it grabs the id from the cookie and requests the personal details from talkr.net (this includes name, email address, url, buddyicon (all, in turn, from flickr), and any notification preferences).
5. If mt-talkr.cgi doesn’t find a cookie it forwards the request to talkr.net and checks for a cookie there.
6. If talkr.net finds a cookie it returns the information stored to mt-talkr.cgi, who sets a cookie
7. If talkr.net doesn’t find a cookie it sets a few cookies (so it knows where to return to eventually) and forwards the request to flickr. They click a button to authenticate, and then travel backwards to talkr, which asks them which information they’d like to allow the blog they’re commenting on to view, and whether or not they want to be notified of new comments, etc. Then it sends them back to mt-talkr.cgi which sets a cookie and uses javascript to add the requested details into the form of the parent window (where they were leaving a comment) and close the popup.

Example use cases:

Seems like a lot of work, but if you think about it most of it happens behind the scenes. The first time you use this it will require the following clicks:

1. Click button to open popup.
2. Authenticate on flickr.
3. Choose which details to allow them to use from talkr.
4. Write your comment and submit.

And from then on, on that particular blog:

1. Click button to open popup and pull your information in (it does everything it needs to and automatically closes after updating the comment form)
2. Write your comment and submit.

After the first time on any blog anywhere, it will take these steps:

1. Click button to open popup.
2. Choose which details to allow them to use.
3. Write your comment and submit.

If you want to change preferences or anything else, there will be another link in the comment form that lets you do that.

You can also go to talkr at any point and set up other things like your PGP signed key, your notification preferences, etc. It’ll also be able to aggregate things and find “most popular posts your friends are commenting on” and provide rss feeds galore.

What happens when a comment is placed on a talkr-enabled blog?

Movable Type’s plugin will notify talkr.net of new comments, and that will check to see who wants to be notified of comments (either by that person or on that entry or on that blog) and notify them (either through email or by updating an RSS feed or by waiting to batch it into a daily email). Easy as that.

What else could talkr do?

It could check to see if the comment is being placed on a trusted public blacklist. It could store your foaf profile. It could remotely submit an entry to your weblog API of choice to build a “recent comments elsewhere” blog. It could convince GWB to step down as President and let bloggers rule the world (wait, that would probably suck).

Am I smoking something or is this super cool?

Sometimes I feel like I’m weak willed. For most things, it’s really easy to change my mind (if I can even find it in the first place). In college, K and a few other friends made me an “Indifferent Man” costume, with a big “I” for indifferent sewed on front. I think it had a butt-flap too for going to the bathroom without having to take the whole costume off.

K often asks me what I think of something, and even though I sometimes give an answer of some sort, internally more often than not my answer is “I don’t know.”

It sometimes goes like this, “How can you not know if you like Super Chocolate Fudge Monkey or the Chocolate Chip Cookie Caramel Banana Experience better?”

I search my brain for an answer, and, finding nothing, say “I don’t know. I just can’t tell which one I like better.” Sometimes I say, “There are too many variables. It depends on my mood, what I had to eat recently, etc,” hoping that will stall the inevitable.

“How about right now? Right now,” she points down to the tubs of ice cream laying like decapitated heads in aisle 13, “at this very moment, which one do you like better?”

I search my brain again, she having called my bluff. “I still don’t know. Should I make something up? You decide. Which one do you like better?”

And then she picks out Super Chocolate Fudge Monkey and we have our ice cream for the night. I realize this is extremely annoying behavior and so sometimes I just flip a coin in my head and say “Chocolate Chip Cookie Caramel Banana Experience is definitely better.”

Then she says, “Really? I like the other one better.”

Then I say, “Okay, let’s get that one then.”

So, over the years, I’ve come to believe that I have a weak will. That my opinions are like drawings in the sand, written by an unsure hand, easily blown away and replaced with a completely new will.

K on the other hand has a very clear and strong will. And so, after a few years of marriage, it’s very difficult to tell what my will is sometimes. Would I take a shower at night (rather than in the morning) if there was no pressure to choose K’s preference? Would I have the same haircut? Would I go grocery shopping more or less often? Et cetera.

Not that any of these answers would be at all interesting. At least, not to me, which is why I don’t have pre-made opinions about those things in the first place. What’s interesting to me is that I don’t know what I would do. If personality is defined by trifles, what is mine?

The other question: is this really a weak will? Is the absence of will a sign of weakness, or is it a sign of perspective? I was recently doing some research on the study of analyzing surveys and an interesting quote came up:

Scaling is the assignment of objects to numbers according to a rule.

Where scaling is the exercise of putting statements on a scale of some sort—Strongly Disagree to Strongly Agree, for example. It can really be said that all opinions are a form of scaling. When most people are asked a question, they are applying some set of rules to the question in order to come up with the answer.

• Have I answered this before? If so, will the previous answer apply again?
• Who is asking me this question? What is their agenda? Should I be careful about answers that might give the wrong impression?
• How will the various answers reflect on myself?
• How do I want to be perceived with my answer, and do any answers help that perception be achieved?

Maybe not everyone goes through all these steps with every question they’re asked, but it seems like it’s pretty easy to shoot right through them pretty quickly and say something immediately. And of course, when someone you know really well is asking you the question, it’s not necessary to think as hard about them—just give the most honest answer, the one that comes to mind first.

But what if some of the rules in your system is:

• What does the asker like?
• How much do they like it?
• Would it be nice to let them have the answer that they like, since it’s really not an important question?

Then what will come to mind first? Their answer, perhaps? Consider this question: “What are you getting me for my birthday?” Isn’t it important to consider what they want before you decide what you want to give them? Gift-giving is an art of finding a solution that satisfies two wills (what they want and what you want to give), and it requires that you know the set of their possible answers before you can start thinking of the set of your possible answers so that you can find the overlap. Where is the line between that and, “What should we eat for dessert?” And “Which movie do you want to see?” And “Are you driving to work or walking?”

Every question in some way involves the asker as well as the person being asked. Am I weak-willed or simply hyper-sensitive to the gift-giving nature of being asked questions? And is that even worse than being weak-willed?

A very stupid title will help me get this out of the way. I’m flipping the switch to Movable Type. Bing! Or more like crank, crank, crack, squeeeaak! I don’t know how many times I’ve seen that phrase (“Rebuilding Erik Benson”) the last couple months. Certainly over a hundred times this last week alone, as I’ve been trying to put the finishing touches on this migration to Movable Type. One more thing, I kept telling myself. Just one more thing. Oh, and then one more. 48 hours later, I can still think of half a dozen more things (one which is truly difficult to fix and which I have been putting off since day one, but hopefully nobody will notice… I don’t think you will).

It has been fun though. I’m writing this, my first non-Salieri-powered entry, in Ecto, which is quite nice except for the fact that it seems to take up a large chunk of my server’s CPU each time it saves a post. What I like is that it’s a nice big text box to stretch out in. I’ve had a lot of Mountain Dew today. And some old pizza. Why leave the house when you can work on migrating your weblog? I ask you that.

So… rather than talk now about what I did, why I did it, etc, I’m going to let you explore or ignore for a bit and wait for things to stop breaking. If you have any thoughts about the changes, I’d love to hear them. Positive and negative. In the meanwhile, I think I’m going to take a walk.

A sign that social networking sites are evolving. Flickr allows you to really easily share, save, annotate, and trade photos with your friends. I’m not necessarily their prime target since I don’t spend much time on chat and don’t take many pictures, but even I can tell that this is the best parts of friendster, orkut, etc, combined with a reason to be. No post-friend collecting depression! There are plans to be able to share other things… streamed MP3s, etc. I’m most excited about the possibilities of their services API that lets to do remote authentication, and some basic other stuff. I’m going to try to use it for my authentication service in my new Movable Type installation (still working on it in my spare minutes), though not sure what snags I’ll run into along the way.

In this case, I feel like I’m as much amazed by the application as with the people who built the application. I guess the completely objective analysis would separate the two, but I can’t help but be really impressed by Ludicorp and their quick and thoughtful development, their lack of trying to copy the existing social software apps out there, their humility and lack of slick and creepy self-promotion, etc. The execution is beautiful, the intentions are honest, etc.

Also strangely sad for some reason. Could be my brain shutting down.

Buster Butterfield McLeod posted a photo:

Buster Butterfield McLeod posted a photo:

Buster Butterfield McLeod posted a photo:

Buster Butterfield McLeod posted a photo:

Buster Butterfield McLeod posted a photo:

face down outside barca

Buster Butterfield McLeod posted a photo:

falling asleep on the street

Buster Butterfield McLeod posted a photo:

Who wants to be my roomie at South by South West? I have a double at the Hilton (the main hotel) but can’t afford to pay for the whole thing myself. Last year I went with Bill and it was quite fun, but unfortunately he can’t make it this year. Be my roomie! For more info, email me through that contact link in the left nav.

So here’s the long-term goal I’ve settled on, with a few further details to help clarify it:

Build reputation, avoid credit

Reputation, of course, is built on credit, so it’s a somewhat counter-productive goal to avoid credit. I’ll have to find ways to build self-explanatory reputation that doesn’t require any self-promotion. I probably won’t build as much reputation as I could have otherwise. It could be said that I’m avoiding effectiveness as well as credit. The problem I kept running into was that most goals, in themselves, are quickly turned into games. Reputation was one of the games that sat at the core of most of the goals, so I decided to abstract it out and let the smaller goals specify ways to actually build that.

By deciding to build reputation I’m making a big sacrifice—I’m putting my goal’s achievement into the hands of others… not ideal but I couldn’t come up with any goals where I could be a reliable judge of my own success… plus, it will help me avoid becoming too self-righteous if I actually have to depend on others on some level. The reputation I’m trying to build can be highly localized to a small group of people, as long as the local reputation doesn’t have a negative impact on the global reputation (for example, I can’t try to achieve this by gaining the respect and admiration of serial killers).

By avoiding credit I mean that I need to avoid building credit as a currency of its own. There are ways to get credit without doing things that deserve credit (or as much credit as you get) and that’s what I’d like to make sure doesn’t happen here.

I told myself earlier that I have to know why I chose the goal that I chose. A goal needs a system... I’ve determined that much. A system is an abstraction of the real world, and it requires that you accept some things as unconditionally true without proof... such as that reputation is worth building. Reputation is a good system to participate within because it already has self-correcting mechanisms to root out selfish and harmful play. And it’s already strongly policed. I could probably write up some rules that would more accurately describe exactly what I want to do with my life, but there wouldn’t be any forced evaluation based on those rules. In other words, I’ve decided to go with the cheaper third-party solution rather than build the perfect system in house. Now I just need to adapt the reputation system by adding one additional rule about avoiding credit, and we’re good to go.

There are many ways to build reputation. Here are the ones that I consider to be candidates for the next 1-5 years:

General:

• Help your friends first, strangers second, and enemies third, avoid asking for help
• Build complex tools that improve relationships between people (make interactions between people more valuable without making them more costly), avoid building complex tools that can’t be used to build/inspire more complex things
• Build health, avoid image of health
• Build strategies for effectiveness, avoid gaming the system
• Build tools for effectiveness, avoid possessions
• Build competitive spirit with self, avoid competitive spirit with others (in case their goals aren’t in line with my own)
• Build large goals directly off of main goal, avoid small goals that have questionable connections even if they are easy to measure

Specific:

• Build a company, avoid money
• Build a book (a real book), avoid publishers
• Build a website/web service, avoid maintenance
• Build an object-oriented written and spoken language, avoid ?

When possible, I’ve added what I think is the best thing to avoid in order to help avoid credit. It doesn’t mean I plan to forsake money, maintenance, and all possessions… but rather I don’t trust myself to seek them explicitly as they are the most common traps that will lead to seeking effectiveness through less-than-honest means. By saying that I need to avoid them, I’m just making sure that they’re not the primary motivator because usually they are very tempting as primary motivators.

The tough part about these goals is that because they rely on people outside myself for validation, it’s really tempting to zero in on a smaller goal and then forget about the bigger one. The end goal is not effectiveness, it is reputation. Many highly reputable people weren’t very effective. Wittgenstein, Salinger, Joyce, and Beethoven built reputation but part of that reputation is in their ineffectiveness—and as a result their reputation is a bit cleaner, more pure, than people like Picasso, Madonna, and most CEOs. Part of my worry is that there aren’t many examples of ineffective businessmen with high reputations. Business and politics seem to be largely dominated by effective people who strove for credit. So that might mean that building a company isn’t an option—I would have to sacrifice my goals in order to succeed, but in succeeding I will have achieved nothing.

My thoughts still aren’t completely clear on this… I feel like I haven’t completely reduced the problem to a satisfactory solution, but it seems to be mostly right for now. I can drop the subject for another year until my next identity crisis meltdown comes.

With this new goal, it would be nice to build tools that could measure my progress along this goal. How to reduce reputation to a relative number? I have some ideas, and they involve social software applications, but none that currently exist. In some ways, reputation can be measured by finding out the number of people who are willing to help you with a certain task. Something as simple as “number of people willing to help you move to a new apartment” might actually be directly related to your reputation. What about building a site called helpfromyourfriends.net or something, which would allow you to tell people what you’d be willing to help them with, and also find out what people are willing to help you with. One interesting side effect is that the relationships become provable. If 10 people say they are willing to help you move, and it comes time to move, you’ll be able to call on those people to help you. Those that don’t might not be able to remain on that help list. It brings the gaming-aspect of the social network into a falsifiable context. Anyone want to build it?

Not many at all. I’m lucky Canada and Mexico are so large. I did the thing for number of states I’ve been to but who really cares about that. The states are pretty lame, except for a couple cities.

Here’s me:

Here’s K:

Ouch. You can get your map here. Follow the meme here.