Infobitt's Future, and Mine

I've just posted the following announcement to the big Infobitt mailing list.



I have some unfortunate news. While I don’t wish to give up on Infobitt, we have run out of money. I’ve let the programmers go, and I’m looking for a job myself. But I'll still be contributing, and I hope you will too.

Before I say anything else, let me say thank you to the investors, my advisers (especially Terrence Yang), and especially the contributors. Thanks also to Vivy Chao, who has written the daily updates very well; Tim Chambers, who provided the awesome audio editions; and Ben Rogers, our technical adviser. And, of course, the readers!

Infobitt deserves to be rescued. It’s got an active, committed community, it’s an awesome idea, it works quite well at a small scale, and I'm confident it can be made to work at a large scale. So we’re very much open to new opportunities for Infobitt. Maybe you can help? I’ll explain how below.

Contents of this mail:

• If you keep at it, so will I
• What’s the core problem?
• Why I'm still excited about Infobitt
• What does Infobitt need?
• Potential partners
• How it can happen
• If not Infobitt: gigs I’d like to consider
• Idea 1: write Philosophy for Children and create a complete set of free philosophy videos for kids to go with it
• Idea 2: making educational videos for little kids—a free online preschool

Please do continue contributing to Infobitt!

If you keep at it, so will I.

If you continue to support Infobitt by writing bitts, adding facts, and so on, then I will too. I do hope that in the next few weeks or months, we’ll re-emerge, re-invigorated, with a new configuration of people who can really make things happen fast.

What’s the core problem?

1. Why don’t we do a proper launch? Because the software works OK only at a small scale. It desperately needs certain features if we are to benefit from the massive traffic we’d get after a proper launch. If we launched now, we simply wouldn’t be able to absorb the new arrivals. (That’s what happened after my Reddit AMA.)

2. So why don’t we just code up the features we need? Because our outsourced software is buggy, complicated, and lacks automated tests, all of which means it’s hard to maintain, and would become more so as we add more (badly needed—see below) features.

3. So why don’t we just raise the money? Because we’re out of money, which makes fundraising very hard. Besides, we need an active, productive team to raise money, and at this point it’s just me, a sole founder.

4. So why don’t I get some co-founders? Yes, just my on.

Why I'm still excited about Infobitt:

• Unlike every other news startup I know of, we are actively, daily creating a purely volunteer, Wikipedia-like front page news site. Infobitt works as no other crowdsourced news startup does. It's been working, in its current version, for about a year now—really working, even if our traffic numbers are still small. That can change (see below).

• People are still working on it, and not just a few, but over 25 every week, and that's on an obscure project that still hasn't been properly launched and is rarely discussed in the media. Regularly, I see old hands getting excited again and new people getting into it. We are onto something.

• I absolutely love your loyalty and I don't forget the people who have helped my projects. You are the lifeblood of Infobitt.

• I've seen evidence of deeper support for Infobitt from outside our active community. There are people waiting in the wings, waiting for the software to get better, waiting to be able to share their work, waiting for it to get easier (e.g., a browser plugin to add facts by selecting text on a page and pressing a button to add to Infobitt), etc.

• When I work more on it, you do. If I were enabled to work full time just on growing the community—if I had the time to write 10 bitts per day, comment and add facts, do more tweeting and blogging, and especially if we were launched and I could do interviews about it, then the community would grow like gangbusters.

What does Infobitt need? So...why aren't we there yet?

• We need a better API. (Our automatically-created Python/Django API lacks many features, although it works.)
• Then we need apps (which use the API). (But a high school kid has actually made one based on our existing API, but it’s not released yet.)
• We need to add some insanely obvious features:

• Fact editing!
• View counts!
• Choose a bitt's rank from within the bitt!
• Social sharing!
• FB/Twitter login.
• Email notices.
• Automatic newsletters.
• Tags/categories.
• Browser plugin to start/expand bitts quickly.
• We've also got serious bugs to fix.
• Any one of these would inject new life into the project. All of them would make this a popular and growing website, I think.

• Then, we need to be properly launched.
• We've got to make the software faster and more resilient for when high traffic arrives.
• I’ve got to start doing interviews. But first we need to be positioned to benefit.

To be brutally honest, I never should have tried to start a startup as a sole founder. I need others on board as partners, who are passionately committed to our mission and to making it a success. I'm doing too many jobs at once, when my forte, what I need to be focused on, is community and project development.

Potential partners. I assume that many of Infobitt’s best potential partners will be reading this, or will know people who are reading this—and you can forward this mail to them. Here is what we need:

• Awesome engineers: Python/Django, Javascript/Angular, PostgreSQL. Solid sysadmin type skills, including experience on AWS, would be most welcome. Somebody who can improve our API so people can make full-featured apps around our (open content) data. Maybe more exciting would be somebody who is inspired (and, of course, positioned) to write Infobitt from scratch, in a more reliable form.

• Designers. (But we need engineers on board first, to be able to use design work.)

• Maybe eventually one or two community people to help me.

How it can happen. Here are some categories of people or organizations who might be interested in joining me and helping to turn Infobitt around:

• Remarkable individuals, especially those are free to work for equity or who might want to buy into the company. Especially awesome engineers who are on top of Python/Django, Javascript/Angular, PostgreSQL, sysadmin, AWS.

• Existing startups, or idle startup teams, that want to pivot to Infobitt, who are interested in working with me. Again, free to work for equity or who want to buy into the company.

• Big nonprofits or fast-moving universities (ha ha). Theoretically, we could become nonprofit, open source, and open content. This would probably make it easier for Infobitt to succeed, assuming the project funding were adequate, but Infobitt's investors obviously would like to make money.

• An investor that wants to buy Infobitt, build a team, and will hire me (with significant equity) and assign me to work on it.

Such people (or entities) would have to buy a major stake in the company and, presumably, hire me as an employee. I’m cool with that.

As far as I'm concerned, everything is on the table. I’ll be interested in anything that has a reasonable chance of making Infobitt a success.

Other gigs I’d like to consider

If nobody bites on Infobitt, here are some opportunities that would intrigue me:

• Full-time worker on somebody else’s startup. Community leader, project manager, or you tell me. I’d prefer to work from home most of the time.

• Adviser. For the right sort of project, I can help a lot. I’m an endless fount of ideas and very useful critical feedback.

• Writer/analyst/advocate. About education, homeschooling, very early reading, the Internet, rescuing the Enlightenment, philosophy, etc. (from a libertarian, rationalist perspective, if relevant). I’m also a practiced public speaker. I’m interested in working for a nonprofit advocacy group.

I'd be excited to execute either of a couple ideas I've had:

Idea 1: write Philosophy for Children and create a complete set of free philosophy videos for kids to go with it
I started writing an intro to philosophy for elementary students, a chapter book, back in 2012. Here’s the first chapter. I’d love to finish it quickly, and use the text to make the world’s first complete set of videos about philosophy for kids approximately 5-10 years old. Here’s the first video. It would take about three months for me to finish if I work on it full time.

Thing is, to support this project, I need at least $17,500. I’d love to do this and make the next generation a bit more hip to the liberal arts and the Enlightenment. I started designing a Kickstarter about this, but I haven't finished it.

Idea 2: making educational videos for little kids—a free online preschool
Are you a philanthropist? Want a high-impact way to support online education for kids everywhere? Pay me me to make 2-3 videos per day like these. Most of those 24 videos got over 10,000 views after a few years, and my top ten have over 50,000 views apiece (with one at 750K). They’re easy for me to make, I’m good at it, and I love to do it. Also, my 4-year-old will beta-test for free! I envision a library of thousands of videos like these...think of it as an awesome free online preschool. By the way, if you want to pay me per video, to make sure I don’t waste your money, let’s do it!

Please continue contributing to Infobitt!

All the best,

I kept my "no social media during work" pledge just fine

As I wrote in my last blog post,

I’m pledging to abandon social media networks when I am at work, except for narrowly defined work purposes. And I’m asking you to hold me to it and slag me mercilessly if you catch me at it! And I’m inviting you to take the pledge, too!

Yep, so for one day at least—and for many more, I still intend—I didn't do any social media at work. I could have done some related to work, but I didn't have any I wanted to do, so I didn't.

I've had a tremendously productive day so far! (Among other things I promoted a plan to get people to write a bunch of one-fact bitts quickly; and I also started a list of our "beat writers," six listed so far, under the first question of our FAQ. Sorry, you may have to log in in order to see this.)

But, sadly, nobody, not even a single person, took this "No Social Media at Work" pledge. Oh, well! I'll continue myself, anyway!

Important Stories Are Hard to Find (Part 2: 4 Reasons the World Needs Infobitt)

I'm sharing four different reasons why the world desperately needs Infobitt. The first was that we have a right to edit the news—that hard, front page news needs input from "we, the people." And now:

Reason #2. The most important stories of the day are hard to find. Look at,, even I frequently find myself disagreeing with what they have as their top stories and disgusted at too much clickbait, emotionally manipulative stories, and other time-wasters. It's even worse on primetime news broadcasts. (So, I'm agreeing with Bill Maher.) And then of course there's the fact that newsrooms are generally limited in what they can cover, and they're going to lead with whatever story they're most proud of or they think will get the most clicks. Of course, they won't lead with a very important story if they aren't covering it themselves.

Aggregators like Google News do a better job of hitting all the highlights than the news sources themselves. But even there, there are problems. Some stories—the admittedly important ones—stay up for a couple of days, long after we've heard about them. Also, sometimes, the big stories of the day are far down on Google News one day, only to make it to the Top Stories the next. Journalists are a special breed (for whom, again, I have great respect); what they choose to write about sometimes represents a judgment, or taste, that many of the rest of us don't share. Many journalists indulge in sensationalism; they seem to be drawn to dramatic stories which have little to no chance of making any difference in the world. The stories are just dramatic. Too often, our time is wasted—and that's true even when journalists' work is aggregated together.

Infobitt works differently. It matters how how many people participate, but it doesn't take many. When we have over, say, 15 bitt writers and 30 people ranking the news, then in my opinion we typically do better than the aforementioned sites when it comes to picking and ranking the stories that matter. These numbers shouldn't be that surprising. There aren't that many "top stories" candidates at any moment; based on my experience on Infobitt, I estimate the number is not much more than 50. Then how many people does it really take to summarize the basics of that number of stories, and to rank them in order of real importance? Again, not that many.

With on the order of 100-500 active daily users—still not many—we'll produce something that has never existed. We'll have an exhaustive, non-redundant, beautifully ranked set of news stories, complete with excellent and well-fleshed-out summaries from many sources, updated all the time.

Already the news on Infobitt is usually pretty fresh. We aim now to make it fresh all of the time and to have covered all the main stories in a timely fashion. Do you want that? Join us and help make it happen! If we haven't covered an important story yet, please add it. In our system, it's called a "bitt." And we always need people ranking new bitts, giving their opinion of whether a story belongs in the top 10 in Top Stories (or a category of interest).

4 Reasons the World Needs Infobitt

This week, Infobitt will welcome thousands of new members (people waitlisted following my Reddit AMA). So, in the coming days, I'll be sharing a different reason why the world desperately needs Infobitt.

Reason #1. We have a right to edit the news. We have a voice in government—we also deserve a role in the Fourth Estate, one we don't currently enjoy. We desperately need a way to make our voices heard about how the news is prioritized and presented.

There's no reason for journalists to be the sole or primary deciders of what we all should know.

Journalists—and I say this with respect for my several friends who are journalists—are experts at some important functions:

• Knowing where and how to find the most important stories of the day.
• Writing quickly and yet readably.
• Forming trusting relationships with newsmakers.
• Summing up the basic facts about a complex situation fairly accurately from scratch (this looks easy but is extremely difficult, and journalists are a whole lot better at it than most of us would be, if we tried).

Journalists serve a crucial and deservedly respected function in society. But they're not any better than other reasonably intelligent, well-informed people at:

• Forming a wise judgment about what the most important stories of the day are.
• Understanding what the hell is going on (there are great reporters who are experts in some things, but most of them aren't anything like experts on what they normally report about).
• Telling you what to think about the news.
• Avoiding bias and corruption in articulating the news.

As reporters of basic and important news, journalists serve an absolutely crucial role in society. But as news presenters and editors, journalists enjoy certain roles that also properly belong to us—to "we, the people."

Simply having reported on some parts of the news doesn't give a journalist an expert perspective on the whole of the news. There's no such thing as an expert perspective on something so vast—"the whole of the news." NBC anchor Brian Williams (just for example) is no better a picker of the top stories than any reasonably intelligent, well-informed person presented with the same breadth of stories. And the idiosyncratic judgment of Mr. Williams and his producers is certainly not better than the average of all of our choices.

Deciding which stories matter and deserve to be placed first is a deeply political one. That decision does not deserve to be made by elites handing down the truth to us lowly plebeians. Top journalists in particular are very powerful: they shape how society thinks about what's going on. They can drive political agendas, boost politicians, make and break reputations, foster revolutions, affect economies, even change our personal habits. And they have important relationships with some of the most powerful people, governments, and corporations on earth.

Only a few journalists are the big decision-makers who determine which stories to highlight and which to tank, of course. But rank-and-file journalists also make important decisions, too—about what facts deserve to be placed in the headline and first paragraph of a story, and which deserve to be buried or ignored altogether.

Before the Internet, it was simply impossible to give us all a seat at the table when it came to ranking news stories and facts within stories. But now it is possible. Infobitt gives you precisely that ability: to rank stories and to rank facts within stories. We're empowering people with editorial functions that they have never had before. The result is a readable, interesting, genuine, and a really useful summary of the news.

Just think: what happens when many thousands, or even millions, of us go to work on it?

So that's one reason to get busy on Infobitt. You're both exercising your own right to occupy the Fourth Estate and supporting the rights of others to do the same.

What do you think? Do "we, the people" have a right to occupy the Fourth Estate? Should we help determine what order stories appear in, how the facts are represented, and what order facts should go in?

Or should we leave these crucial functions to the professionals?

Please return tomorrow for reason #2. For more blogging about the project, please see the manifesto.

The top 10 things for contributors to know about Infobitt (that aren't obvious)

So you're interested in contributing to Infobitt.comeh? Excellent! Here's what you need to know.
1. You gotta understand our mission. If you don't get it, you won't be motivated. Read about our mission further down in the FAQ. Hint: our aim isn't just to make "another citizen journalism site." That would be boring. Our aim is to organize the news. This will take an enormous online community, a movement, like Wikipedia's, if it's going to work. What we have is already cool, but it will get cooler as our community and software grow.
2. Contributing is easy. Click + to add a bitt. Rank bitts by dragging and dropping them (on a computer, not a handheld, for now). Please add facts to undeveloped bitts: click in the "Add fact" field and start typing. You can do it!
3. The first fact should not be a title. It should be a complete, grammatical sentence. We're writing summaries.
4. Be bold. Don't be shy! You have the right to participate and we want you to!
5. Rank bitts in order of importance as serious news items, not "coolness" or perceived popularity.
6. Avoid duplicates. Use our snappy search feature if you aren't sure we've included a story yet.
7. Leave comments! Voting/ranking is nice, adding facts is nice, but sometimes we need to coordinate and discuss stuff if we're going to make this project work.
8. Participate a modest amount regularly rather than a whole bunch all at once. The news doesn't happen all at once!
9. Please don't start more than five bitts per edition (12-hour period). More than that discourages other contributors.
10. Don't let the news get stale. If all the bitts on the site are over 6 hours old, and especially if they're over 12 hours old, trust us: there's a lot of newer news we aren't covering. Our goal is to have well-developed bitts about all the latest developments, as soon as they happen! Please help! Also, see our help video.

How we can organize the news (short version)

This is the first public discussion of Infobitt.

You can now sign up for an account without an invitation. We're starting a 100,000-person pledge drive: when we reach 100,000 pledges to add one fact, we’ll ask everybody to show up at once! We’ve also opened up our daily newsletter and have kicked off a discussion group. If you join, be forwarned, it's a little buggy; you might have to refresh. But do join us!

There's a long version of this essay.

I'm co-founder of Wikipedia. Now think back to a time before Wikipedia—the 1990s, if you’re that old. If you didn’t know the answer to a question, and a web search brought no joy, you might have to take a trip to the library, or stay ignorant.

Today, if you don’t know the answer to a question, you can find one on Wikipedia within seconds. That’s a stunning development for humanity: we now have virtually instant access to answers. That’s a historical first. It changes how we learn, how we communicate, and how we think.

How did it happen? Millions of people from across the globe understood the vision of a free, open content encyclopedia and acted on it. It was my job to organize this effort. Wikipedia was the result.

Now I hope to organize people to summarize and rank the world’s news in a free, open content news resource. The project is called Infobitt.

It is human to want to know what’s going on. Perhaps billions of people follow the news every day in one form or another.

This is why there are endless streams of news flooding the media landscape, more than ever before. Many of us would like to stay on top of the spreading pool of news. But the flood is getting deeper and wider.

That’s why we skim various news sources, apps, our friends’ feeds, blogs, etc. We resign ourselves to not knowing many details; for that, we must read many articles in full. Only devoted newshounds have time for that.

Beyond the sheer quantity of news, we must navigate redundant news, click bait, sensationalism, and on and on. The news has become noisy and confusing. There’s a sea of it, and it’s uncharted.

The facts about an ongoing story are often spread across many different sources, from the New York Times down to a humble blog. Nobody organizes the facts. No media outlet has the motive or the ability to come to grips with everything.

But we do—billions of us have that motive, and if we are organized in the right way, we’ll have the ability. What if we pooled our efforts on the news in the way we did on Wikipedia?

But how?

Here’s the Infobitt model. We grab different facts from different news sources, summarize them in sentences which link back to those sources. We each drag-and-drop the facts into our preferred order, and the system calculates the sense of the community. The result is a bitt.

That’s not all. There’s a stream of new bitts arriving in the system. We put bitts in order of importance by drag-and-drop as well. We’ve made a new way to collaborate on collecting the news.

We want this done for every article about every story. And we want it constantly updated. After all, it’s 2014.

Only a giant, international, online community could make this happen. This is citizen journalism re-envisioned to include an enormous distributed editorial function. It’s ambitious, but we can do it.

Here are the Top Stories from late October:


And here’s a representative bitt:


The back end of the software isn’t simple; it has lots of interacting rules. To fund its development, we raised a seed round a year ago. The system, now in its second version, is easy and fun to work on. No invitation is needed sign up, but be forewarned, it’s under development; refreshing fixes some bugs.

A small group of beta-testers have been using it for several months, and they regularly produce bitts and editions (especially the evening edition; subscribe here) of surprisingly good quality. We’ve opened a mailing list to discuss project policy.

With just 10 times the number of contributors, the top stories would always be fresh and detailed; that wouldn’t be hard. We can do this!

We’ll have to squash bugs and slowdowns fast enough, of course. We are looking for A-list web coders, so send us leads to awesome programmer who might be excited to work on a startup like this.

We’re announcing a pledge drive. I’m asking you to pledge to add one fact to the system—just a sentence and a source URL.

When we reach 100,000 pledges, then we’ll contact all the pledgers, and we'll say, “OK, people, let’s get this party started!” Everybody converges on the website for 24 hours of fun, and our project really kicks into high gear.

People used to tell me that Wikipedia would go nowhere. They were wrong. Infobitt too tackles a universal problem and requires an ambitious solution. We’ll inevitably have naysayers. But they’ll be wrong, too. The latent demand for this is strong, and a lot of people will naturally want to organize the news. We can do this.

It's potentially revolutionary whenever a bunch of smart people from around the world pool their efforts in making sense of information. We’ve done that with general knowledge. Now let’s do it with the news!

What do you think of the idea? Let's discuss below.

How we can organize the news (long version)

This is the first public discussion of Infobitt.

We did a soft launch recently, meaning you can sign up for an account without an invitation (but that’s the only way you can see the site, for now). We are starting a 100,000-person pledge drive (when we reach 100,000 pledges to add one fact, we’ll ask everybody to show up at once!). We have also opened up our daily newsletter, and we are soon to kick off a discussion group. Please bear in mind the site is a little buggy; you might have to refresh. But do join us!

There's a short version of this essay.

I begin with personal background. You can skip straight to the part about Infobitt.


Part 1: Background

Dreamer, nothing but a dreamer

I am a dreamer.

That might sound a little pathetic and weird for a 46-year-old short, bald guy to say. But ask anybody who knows me well, and they’ll agree. My life has been a never-ending succession of dreams, a few of which have come true.

In my childhood I dreamed of being a cartographer: I used to draw very detailed maps of my home state of Alaska from memory.

As a teenager, I dreamed of being a novelist, until my brother told me I was 16 and therefore had nothing to write about. So I decided to study philosophy, so I would become wise and have something to write about.

That led to a dream of being a philosopher and writing a great system of philosophy. So I majored in that.

While I was a grad student in the 1990s, I dreamed of playing the music I liked to listen to—Irish traditional—and took up fiddle.

I remember a dream I had in 1993, and I know the year because it was the year after the billionaire Ross Perot lost the 1992 presidential election. See, I wrote to Perot with this idea of getting a bunch of people together to summarize the news. I was excited, but of course it was half-baked and silly. I was only philosophy student and had no clue about how the business world works.

Another thing I did back then was start to organize online communities, in the form of Internet discussion groups. One of them was devoted to the idea that people could tutor each other online, leaving bloated, far-too-expensive universities behind. That was a dream and still is. (Many years later, this idea started getting real traction.) There were also a couple of philosophy groups, one of which (the Association for Systematic Philosophy) Wikipedia co-founder Jimmy Wales joined; I had joined his group MDOP. These groups all taught me that people working together on common interests could produce things of great value for everyone involved, things worth spending many hours on.

My dream of becoming a professor withered. I still wanted to be a philosopher, though. I remember thinking about becoming an encyclopedia editor. Not sure why I didn’t try harder.

I ended up, in 1998, starting a website (the content appears to be archived) with summaries of stories about the Y2K problem (the Millennium Bug). Jimmy Wales followed the website but we argued about whether there was anything to worry about. I wasn’t sure one way or another; he was a “naysayer.” He turned out to be right. Anyway, the site was popular but, since the world didn’t end on Jan. 1, 2000, it had a limited shelf life.

After that, I came up with another idea for a website. It was the dream that I pitched to Ross Perot, but now I better understood how it could be implemented. What if everyone were to write summaries of the news as I did about Y2K? I knew that online communities could be motivated to work together to create great stuff that they shared freely. It later came to be called “peer production.” So I shared the idea with people I knew online.

Wikipedia days

One of them was Jimmy Wales. He told me to forget about my dream of a free, community-produced news summary site and come and work on his dream: a free, community-produced encyclopedia.

That job offer was a dream come true. I came to work for Jimbo’s company, Bomis, in San Diego. Bomis had various projects, all funded by their main product, a web portal called, and a thriving Internet boom. By the time I had arrived, some ten days later, Jimbo’s dream of a free encyclopedia had become mine, and then some: I knew the potential of groups working together online. I really believed it was possible to get them to write an encyclopedia. There had to be a way. It was just a matter of time and iterations.

Starting this encyclopedia was a great job because, though I often sought Jimbo’s direction or advice on the project, he rarely gave it and specifically and repeatedly said it was up to me. I was able to get creative on somebody else’s dime. Jimbo even gave me a raise when, in early 2000, I finished my Ph.D., fulfilling another dream.

Our first “draft,” ultimately thrown away, was Nupedia. Nupedia was moving too slowly. The model wasn’t right; it was too complex. So it was my job (I was, after all, the full-time leader of Bomis’ encyclopedia project) to find a new way to involve more people.

Thus one fateful evening, Jan. 2, 2001, I had a Mexican dinner at a café in Pacific Beach with my friend Ben Kovitz. He was telling me about his interest in wikis. Now the next part seems to have become slightly confused. I don’t recall Ben suggesting, per se, that we should make a wiki encyclopedia. But he says he was talking about wikis in order to get me interested in the idea of a wiki encyclopedia. I don’t doubt that, and I certainly don’t begrudge him dropping the idea in my head, if that’s what he did: clearly it worked.

What I recall is that at some point in the conversation, I said something like, “This is very interesting, you know. I wonder if wikis could solve this problem I’m working on.” Skeptic that I was and am, but increasingly familiar with managing online content projects, I had a lot of questions about wikis and whether a wiki encyclopedia was really feasible. But based strictly on Ben’s description, I thought it was worth testing out; all we had to do was install a wiki, after all. So I asked Jimbo to have the software installed so I could play around with it.

Coming up with the idea isn’t what got Wikipedia started. The events of the first year did that. As I had done many times before, I started the community by inviting people and interacting with new arrivals. I watched the “recent changes” page, commenting and thanking people for their contributions, formulating our initial rules, writing essays, interacting with people constantly, and promoting the site elsewhere. Getting Wikipedia started became my full-time job and I loved it. (By the way, I’m just telling the same story I have told whenever anyone asked ever since 2001. I’m just telling it again.)

Then toward the end of 2001, the Internet boom turned to bust, as did the market for Internet ads—Bomis’ meat and potatoes. So Bomis had to lay off all but their original few people. I was the last to go, I was told. There was no money in ads, so my position as encyclopedia organizer, just representing a pie-in-the-sky dream for the company, couldn’t be sustained; Wikipedia, though it was growing like gangbusters, just couldn’t earn the money at the time to pay me.

I was getting married, so the situation was hinted at beforehand, and then I was given more official notice when I got back from my honeymoon. If I was then unemployed, at least I was able to live my dream of being married to an excellent woman. I left the project permanently in March 2002, 14 months after starting Wikipedia and two years after starting work on the idea of a free, volunteer-built encyclopedia.

Later I felt compelled to distance myself even further. My dream, or rather my specific version of a widely shared dream, was of a free, open, public-built encyclopedia—but also one that was a pleasure to contribute to and really authoritative. As of 2014, despite being a wonder and extremely useful, Wikipedia is still not authoritative and I still hear many stories of how difficult the community is to deal with. I still think we can do better.

In fact, I tried to do better.


After Wikipedia, I was actually unemployed for a little while. Friends of mine had become programmers, and it looked like fun, so I taught myself Perl. One of my projects for learning the language was a website that would—once again—let people collaborate on news summaries. But nothing came of that; the requirements became too difficult for a beginner to execute, and I didn’t stick with it.

I spent a couple of years teaching philosophy and wondering if any of my old dream of being a philosophy professor was still there. It was not. I still wanted to be a philosopher, but a professor not so much.

So I was hired by a Bay Area startup in 2005 that was going to make an attempt at a new sort of Internet resource, but after a year it was clear they didn’t have the funding to support my plans. One good thing that came out of that is that I was able to help get the wiki-based Encyclopedia of Earth off the ground. So, in 2006, I started Citizendium, another wiki encyclopedia. I won’t tell its story, but I will say that it’s still going, lo, these eight years, because there is a die-hard core of people who still share my dream of a different kind of wiki encyclopedia. I remain very grateful to them.

But ultimately the biggest problem Citizendium faced was the fact that it launched at the exact moment when Wikipedia was growing fastest, in 2006. A lot of people thought Citizendium was a good idea, but ultimately they went back to work on Wikipedia, simply because it dominated the scene and that was where the world could see their work. In a world with a limited number of volunteer encyclopedists, it’s probably true that only one could come out on top.

Citizendium was a nonprofit, and I had to seek funding. An elderly Memphis philanthropist, Charles Boone, started helping, but eventually he invited me to adopt his dream of getting people to make excellent educational videos. WatchKnow, later renamed WatchKnowLearn, was the result.

Around 2008, a dean I think from Duke University contacted me out of the blue and encouraged me to apply for a position that would lead the development of a news crowdsourcing site. I told him I had ideas about that but no time to execute them, not without abandoning my other projects; so I reluctantly had to pass up the opportunity.

While Citizendium and WatchKnowLearn were starting, I fulfilled another dream. My wife and I had a couple of kids, and we’ve later been able to homeschool them—another dream. Also, I dreamed of teaching my first son to read when he was very young, shortly after seeing various YouTube videos of other kids doing similar feats. It seemed implausible at first, but it turned out to be an achievable dream.

When Charles saw a video of my son reading the First Amendment at age three, he asked me in 2010 to finish a book-length essay on baby reading I had started and to create a web-based tool that would digitize the flashcard method I had used with my first son. My second son made use of the result, called, and he too started reading very early.

Reading Bear turned out to be a massive effort (it’s huge), but we launched it in 2012, which was another dream fulfilled. A lot of people love that site, which is 100% free, ad-free, and nonprofit. Charles wanted me to promote Reading Bear, but I decided someone else would have to do that job. So, parting with him amicably, I found myself wondering what I’d do next.

Part II: Infobitt

The problem

In late 2012, I returned to the dream I pitched to Ross Perot in 1993, shared with Jimmy Wales in 2000, coded in Perl in 2003, and sidestepped in 2008. But it was a new approach.

It can be described briefly but somewhat misleadingly as Wikipedia for news; but it's not a wiki.

Think back to the year 2000, before Wikipedia. If you didn’t know the answer to a question, and the answer didn’t come up immediately in a web search, you had to look it up in a book or ask someone knowledgeable.

Today, if you don’t know the answer to a question, you can find the answer almost immediately in Wikipedia—or by searching Google or by asking Siri, which both heavily rely on Wikipedia.

If you think about it, that’s a stunning development for human beings: we now have virtually instant access to answers. That’s a historical first.

Millions of people (all told) from all over the world were ready and willing to record their knowledge. Wikipedia gave them a way to do that. They showed that our dream was a practically universal dream.

Now let’s talk about another deeply important kind of knowledge: the news.

There is practically universal interest in the news. Perhaps billions of people follow it every day in one form or another. It is human to want to know what’s going on. So you’d think that, by 2014, it would be easier than ever to catch up on the news.

In some ways it is. There are more streams of news flooding the media landscape than ever before. Many of us would like to stay on top of the spreading pool of information. After all, we value the productions of journalists; just think of how many of us share neat articles on Twitter and Facebook. But the flood of news content is getting deeper and wider.

In an effort to at least see all the important headlines, the latest way to consume news is to skim many different sources, read news apps, consult your friends’ Facebook and Twitter feeds, watch videos, read some blogs, etc. This shotgun approach acquaints us with more stories than in pre-Internet days, sure—but we don’t understand the details unless we read the stories. Who has time for that? Only the news hounds; the news is their hobby (or profession).

The problem is not just the sheer quantity of news. Speaking for myself, I can’t stand the cacophony of redundant news, click bait, human interest stories, different news angles that add nothing, opinion masquerading as news, news broken by opinion columnists, reheated old news, and so on. Perhaps you feel the same way.

The news has become noisy and confusing. There’s a mountain of it, and it’s completely disorganized.

If you religiously stick with it for a couple of hours every day, then you will be caught up with the news all the time. (I know; I’ve tried.) It doesn’t matter what tools you use. There simply is no tool that organizes all the important news intelligently and filters out the redundancy and other noise.

This is why only the news hounds are really well-informed.

This is also why so many news apps are personalized: they know we just don’t have time to get caught up with everything we’d like to, so the apps help us make choices based on our interests. That strikes me as giving up on getting well-informed. That just makes you informed about your interests, perhaps, but not about the news, period.

Other news apps and emailed newsletters try to make reasonable selections of the top stories for you, curating the news and summarizing it. They are an attempt to solve the same problem, but they don’t: you’re still missing a lot. And you don’t want to.

News hounds of the world, unite!

The news about an ongoing story is often spread across many different sources. High-density sources would include AP, Reuters, and The New York Times. But then there are the second- and third-tier sources, blogs and Twitter feeds of journalists and newsmakers and eyewitnesses and experts, YouTube channels, influential academic journals, and more.

The facts about a story are scattered across dozens of sources. Nobody organizes them. No one of those sources has any motive, or the ability, for that matter, to promote the other sources and summarize everything.

But we do—we, meaning almost everyone in the world who follows the news. Billions of us have that motive, and if we unite, then together we’ll have the ability. So what if we pooled our efforts on the news in the way we did on an encyclopedia?

I have a dream about how we can do that.

Will you consider making this dream your own?

The news is made up of facts, significant facts that have come to light recently, and the facts are scattered everywhere, as I said. What we want is all the different facts about a story picked from different news sources, thrown together in one place, and then ranked by importance. Finally, we want the facts linked back to their original sources.

This is exactly what we want and need. It would save us all a lot of time, because it would make it really easy and efficient to get caught up with the news about a story.

I’ve just been talking about one story. We want this done for every article about every story. We want it up to date at every minute of the day. After all, it’s 2014. We see what online communities like Wikipedia, YouTube, Pinterest, and others have done. We can imagine a community doing it.

But only a giant, international, online community could make this happen. Organizing thousands, and perhaps even millions, of people to work together in this way is the only way human beings could do it.

Highly-paid programmers have tried mightily to do something like this with automated summaries, finding news clusters automatically, ranking them, and so forth. But the fact is that, like writing encyclopedia articles, deciding how to summarize the facts, how to rank them, and how to prioritize whole stories as they break—this all requires careful human judgment. And if you’re talking about ranking and summarizing a vast, never-ending stream of facts and stories, no traditional paid editor, not even a large team of paid editors, is up to the job. It’s a job for a massive online community. Like Wikipedia.

You might say that this is just citizen journalism. Well yes, it is. But I maintain that we have not fully grasped the potential value of citizen journalism. It’s not just about ordinary people adding more facts to the flood of information out there. That is important and potentially revolutionary. But unless there is a way to organize and make sense of the flood of data, the voices of citizen journalists will be drowned out. So citizen journalists must also perform, democratically, an enormous distributed editorial function, one that traditional journalists simply aren’t equipped to perform.

If many of us together ranked news facts in order of importance, then sometimes, a fact sourced only from a blog or a tweet would jump to the top of the list. And they should! Finally, citizen journalists really could have a loud and clear voice if they broke some interesting news.

This is the way, perhaps the only way, to finally shine a clear light on the long tail of news reporting, capturing not just the high-profile professional reporting by the wires and world-class newspapers, but also the humble blogs.

Let’s do this. In some years, we’ll be able to look back and say, “Remember when it took a couple of hours really to get caught up with the news of the day? Do you remember when, half the time, you didn’t even know what was going on? Remember when important facts sourced from authoritative blogs couldn’t get equal billing to those from influential reporters? Remember when you didn’t want to read the news because the noise made it so much work?”

Infobitt’s story so far

So that’s the Infobitt vision. How far along are we to achieving it?

By late 2012, when I first had the idea, I had been working in academe or for nonprofits for almost my whole adult life. Almost all of the nonprofit projects I had worked on had very limited budgets. I didn’t want this project to be spoiled by lack of sufficient funds. I knew that if it were a for-profit project, we would have a shot at raising the money we needed for necessary experimentation and iteration. After all, the software didn’t exist yet, and the idea is very complex and there’s much that could go wrong with both conception and execution. For a really ambitious idea like this, a long runway was needed.

I thought the nascent enterprise might parlay my reputation as Wikipedia co-founder to get funding. So, with help of an awesome early investor/adviser, the L.A.-based Terrence Yang, we did! We ultimately closed our first round in November 2013. I thought the seed capital we raised wouldn’t give us a long runway if we didn’t cut expenses to the bone, so we did. We focused on building a working tool.

Here’s how it works now.

We begin by making bitts, which are collections of news facts. We get citizen journalists, news hounds, and you together into a news-curating community. We grab different facts from different news sources, summarize them in sentences, and put the sentences in order. The result is a bitt. You can add to any bitts in the system, and you get points both for adding facts and for creating new bitts. We also vote on fact order: we each drag-and-drop the facts into the order we prefer, and the system calculates the sense of the community.

That’s not all. There’s a stream of new bitts arriving in the system. We want to be put bitts in order of importance. “Importance” could mean your prediction of what will have the greatest long-term historical impact, your judgment of what most profoundly affects us, or whatever you think “hard news” means.

Editors vote on bitt order by drag-and-drop as well. By “editor,” I mean you, or anybody who shows up to work on the system. Sometimes we call ourselves “bitters,” because we’re writing bitts, but without bitterness—although the news is often a bitter thing.

At any moment there are ten “Top Stories”; the rest of the bitts are filed away in categories. We’ll add a tagging system too.

From the reader’s point of view, the Top Stories list looks like this (for now; it needs a snazzier design, and the content will improve too):


And here’s a representative bitt:


This is actually the second version of Infobitt. We started seriously coding the first version in summer 2013. We worked on it for about nine months, then decided to pivot because it wasn’t easy and fun enough; our early adopters were losing interest. We started coding the second version last March. It’s a lot easier and more fun, so we’ve had more enthusiastic feedback.

Unlike with Wikipedia and Citizendium, the software was not already written when we started, so there was little I could use to attract a community. I didn’t want to try to cause a splash with a mere idea that people have to wait to be developed; talking about vaporware gets old very fast. So for purposes of testing and feedback, I started a beta-tester list with around 80 people (to whom I am very grateful).

Despite being a “stealth private beta,” we’ve been creating 119 bitts per week since August, with 8-20 people out of that small group working on it every week, 4-12 per day. Two editions per day are posted automatically. The quality of the evening edition (subscribe here, it’s just a hand-managed Google Group for now) is to me surprisingly good for such a small number of people. The way the software is working now bodes very well for the future.

The software is fairly stable, but not entirely bug-free. If you join us, please be patient, and remember to refresh if you need to.

We quietly removed the invitation system and are now, for the first time, starting to promote the Infobitt vision and build the community. It’s not hard-launched because the system still has bugs and is under heavy development.

Still, the current version is now good enough to start galvanizing a community around. Do join us!

How we will grow (with your help)

So, you say, this all sounds very nice. But it’s one thing to talk about thousands or millions of active contributors. It’s quite another to cause them to show up. As many sad startups have discovered, it is not the case that if you build it, they will come. So what’s our plan to build the community?

First, Infobitt’s content will be open content, as you might expect: if you contribute, and even if not, you’ll be able to reuse the content from our feeds in your own projects. We haven’t finally decided which Creative Commons license we’ll use, but it’ll probably be CC-by-sa. The point is that our contributors will know that they are building a public resource.

While an encyclopedia project needs hundreds of active contributors to make much headway, we’ve started an interesting news report with far fewer people. With our small group of around 8 beta testers per day, we’ve been able to create a pretty good evening news report on most days.

Here’s why I find this exciting, although I am speculating:

With 15 active users per day, our news would be reasonably fresh throughout the day, with both morning and evening editions being good. Also, most or all the top stories would be fleshed out with many facts and multiple sources.

With 30-50 active users per day, the top stories would be consistently fresh and excellent; the community might demand more editions per day; generally, the site would have the latest news within an hour or two. Also, 3-5 of the main categories would be consistently fresh throughout the day.

With 100-500 active users per day, we’d see news being posted within minutes of its breaking. The top stories would be fully fleshed out with, perhaps, dozens of facts from many sources. People would create “beats” in which they frequently check up on specialized news feeds, Twitter feeds, blogs, etc., multiple times daily, in order to be the first to add new facts, or new stories, to Infobitt. All the top categories would be high quality and rarely lag behind breaking news more than an hour.

With thousands or millions of active contributors, the top stories and the main categories would be a marvel unlike anything seen in the history of journalism. Every story would break on Infobitt within seconds of breaking anywhere else. Bitts would be exhaustively detailed within minutes. For the top 50 stories of the day or so, the sources would be carefully selected and vetted. The writing of the top stories would be as excellent as in the New York Times. The ranking of bitts and of facts within bitts would be superb. Other exciting features would be achievable on that scale as well.

Now, 500 active users on a peer-production site is nothing to sneeze at, but it’s also not ridiculously ambitious to aim for. We can do this!

That’s assuming, of course, that we can squash any crippling bugs and slowdowns fast enough. Bugs and slowdowns seem inevitable. But we’re totally committed to fixing them as fast as we can. It will help when we hire some A-list web coders, so send us leads to awesome programmers who might be excited to work on a startup like this.

I’ve got another idea for growing our community. When Wikipedia got started about 14 years ago, many people still weren’t online, and most people who were online weren’t deeply immersed in the Internet the way they are now. Back then, there were lots and lots of little communities. So I initially recruited many people for for our free encyclopedia projects by reaching out to those groups. But since then, the social organization of the Internet has changed a lot. We really couldn’t organize Infobitt the same way today. Little communities still exist, but now, most people are organized through the big social networks like Facebook, Twitter, LinkedIn, and SnapChat, where things can go viral very fast.

The few-thousand strong grassroots community that it took me over a year to gather in 2000 and 2001 for Wikipedia could be done in a matter of days or weeks in 2014—if we went viral.

I have a plan about how Infobitt can go viral—how we can go from being a cool dream with a beta website to a massive community excited to work together making the world’s first peer-produced front page news site.

It begins, in fact, with this very manifesto.

I’m announcing a pledge drive. To join, you pledge to add one fact to the system. A “fact” means a sentence, preferably in your own words, summarizing some fact that you found in a news article, together with the URL of the article. Easy, right?

When we reach 100,000 pledges, then I contact everyone who has pledged, and I say, “OK, people, let’s get this party started!” Everybody converges on the website for 24 hours of fun, and our project really kicks into high gear.

Of course, the big question is how we collect the 100,000 pledges. This is where I need your help. I want you to take the pledge. Not only that, if you value democratic involvement in news collection, ranking, and summarizing, you could share the pledge drive and the project (a few times—many people won’t get it the first time) with your local networks. I’ll do my best to get publicity for the pledge drive. I’ll do interviews. We’re making a promotional video. I’ll write more.

Within the growing Infobitt community, we’re opening two mailing lists. On one, we’re posting news editions daily (please subscribe!)  and discussing project policy (join if you like!). Soon we’ll set up a wiki where we can hammer out the rules, or guidelines. Just bear in mind that most project policy won’t have the same bureaucratic weight as on Wikipedia, because the content and its ordering is all determined by your voting habits, which needn’t be strictly controlled by the rules.

I can only hope that by that time we do have 100,000 pledges, is in a state where it can handle 100,000 simultaneous users! But wouldn’t that be cool? In our interconnected world, it’s not unusual in the least for things to go viral, and 100,000 pledges for a project as cool as this would mean we’d kick Infobitt into high gear.

Let me briefly talk about project governance. If we're starting a giant news community, do I want to become the dictator of the news? No. I want the community’s governance to be democratic; ultimately I want it to approximate a constitutional republic. Reining in bad actors on the site (a chronic problem for Wikipedia) will be done by randomly-selected users and mediated by automated tools. The brief, modular nature of Infobitt's facts makes this feasible.

Let me make one thing clear. I’ve rallied people to work together on high-minded collaborative projects before. I’ve been doing it since the mid-90s, and I really like doing it; I love the barn-raising sense in us, the sense that we can come together and create something really awesome out of nothing more skill, good will, and mutual trust. So I have always felt that the project’s the thing: I just want Infobitt to succeed. Like Wikipedia and my other projects, it’s a lot bigger than me.

So, to maximize our chances, I’m going to listen to you closely and make sure you’re happy and motivated. I’m going to get it off to a roaring start, and then, as I’ve done many times before, I’m likely to step aside and hand the reins to someone else. Frankly, I’m a starter. I have no interest in being the pointy-haired boss or a community figurehead. Once the dream is alive and thriving, I’ll probably be off working on another one.

Call to Action

We’ve already done a lot toward getting this project ready for wider participation. But we still have a lot to do.

You can help! Here’s how:

Pledge to add a fact when 99,999 other people have pledged to add a fact. We’ll hold you to your pledge! Fortunately, adding a fact is easy.

Create an Infobitt account.

Send Larry an email at with the subject “I’m in” in order to volunteer for a 50-person test pledge day, which we’ll do much sooner.

You could read the news with your new account!

Better yet, contribute! Add bitts, add facts to existing bitts, and rank bitts and facts.

Watch the brief orientation video (a bit old, but still basically correct) and check out the FAQ as necessary.

Give Larry leads to more awesome programmers who want to work on this startup. We have started a search for a full-time CTO

Join Infobitt-Daily-Update for a peer-produced top stories of the day, as well as news about the project.

Join Infobitt-talk and discuss the future of the project. If you’re really interested in this project, you no doubt have a zillion questions. Possibly also a zillion opinions. Great!

A project policy wiki is coming!

One last word.

Remember what we can achieve with relatively few active contributors. We can do this.

People used to tell me that Wikipedia would go nowhere. They had all sorts of objections. The quality would be awful. We wouldn’t be able to handle the vandals. We wouldn’t be able to get people to volunteer to write encyclopedia articles (who wants to do that?).

The naysayers were wrong. We got together and created a resource that enables us to get practically instant answers to very many questions.

Infobitt tackles a similarly universal problem and requires a similarly global solution. The problem seems enormous and unsolvable: think of how much reporting is done every day, not just by mainstream news sources but by blogs, videos, tweets, and other things. Summarizing and ranking all the facts contained in that steady stream of data seems impossible. We’ll inevitably have naysayers.

But they’ll be wrong, too. We can do it.

And in time, perhaps soon, we’ll be in a position where, if you want to get caught up with the news, you don’t have to skim a dozen different sources and try to detect the latest important facts through all the redundancies, deciding what is important out of all the noise that the news media throw at you. You’ll still have to make choices—you still won’t be able to stay abreast of everything—but you'll have a credible starting point that makes more sense.

In 15 minutes you’ll be as well-informed as a news hound was who formerly took an hour to comb through the confusion that is the news media.

As a result, we’ll all be better informed. And tyrants had better beware, just as I used to say they should beware of Wikipedia. When you have thousands of smart people from across the globe pooling their efforts in making sense of information, it becomes more difficult to count on widespread ignorance. We’ve done that with general knowledge. Now let’s do it with the news.

What do you think of the idea? Let's discuss below.