inluu territory

Well, after a summer of challenging health issues, family visits, other projects, resources falling off and trying to have *some* free time, there’s actually a beta version of inluu for you to check out! I had to do a last-minute scope reduction to get the thing launched before the 24th (whose importance I will get to).

Therefore, I have disabled features like public/private memory setting, memory time line association, image uploading, many customizable settings, advanced search components, etc. etc. Another feature I call deep memory association has also been disabled for the time being, which limits the chat responses somewhat (but it’s still really cool). As you can see, the list of features under development is rather long. I guess that gives us all something to look forward to.

Today, September the 23rd, was a hard deadline for me because I have accepted a great full-time opportunity with another company starting tomorrow. inluu thus becomes one of several part-time projects I will pursue in the future. inluu is near and dear to my heart and I promise development will continue as rapidly as time allows (man, do I wish I had a couple of other developers on this).

Rafa, a friend and former manager of mine, once told me: “I’ve seen you in front of computers, man. Other devs will give up and go home to mommy. But you… you beat the computer bloody, threaten it with more violence, force it into submission and make it do what you want.” Well, Rafa and everyone else, let me assure you that there isn’t much left of my computer other than a reeking puddle of plasma.

So, please check it out and remember that since it’s a beta version, data is subject to destruction without notice.

Signing off,

/Juan

One thing you quickly realize when you’re going it alone is that there is no Bob or Fred or Jane to cover for you when you don’t work. Having spent the last month moving, it’s now crunch time for inluu prior to going to beta on September 10, with the launch anticipated a few weeks later.

So, what remains to be done? Well, the boring stuff. The core is complete, so it’s time to wrap up the 70 or so open items surrounding user registration, metrics and other random components. With the meat of the application done, however, I feel comfortable setting a specific date for the beta. Keep your eyes open for an e-mail from me if you’re on the beta list!

/Juan

Brier Dudley of the Seattle Times was kind enough to mention inluu.com on his business/technology blog yesterday. If you’re interested in Seattle area technology news, be sure to subscribe to Brier’s blog.

Also, John Cook of the Seattle Post-Intelligencer mentioned us on his excellent venture blog.

Finally, Outside the Valley today made an interesting post about us. I’d like to thank them for the mention as well.

/Juan

In an unprecedented “let’s make two posts in a day” kind of way, I’d like to share a couple of inluu.com pre-beta screen shots with you. These were rendered by the real deal and hopefully help illustrate what I was babbling about in my previous post.

Training

Chat

Details are subject to change, of course, but overall this looks relatively close to what I expect to launch. I hope to put together an animated/slide demo soon but finalizing the code is a higher priority right now. Your comments, as always, are welcome!

/Juan

Alright, folks. This thing is close enough to beta now that I feel fine revealing the specific workings of inluu.com. You already know that my intention is to help people capture their life stories and memories, allowing them to share both epic moments and trivial things with their friends and family, and indeed the entire world.

How does inluu.com do this? Well, there are two pieces to it:

• A training tool, which you use to teach your own inluu about yourself
• A chat feature, which other people use to chat with your inluu (in lieu of you)

Let’s geek out…

Training

When you first come to inluu.com and decide to train your own inluu, you’ll be asked to introduce yourself. You type your response using natural language, in your own words. Perhaps you’ll say something like:

“My name is Juan Irming and I live in Seattle. I was born in Spain and raised in Sweden. I’m married to Dana and we have a son, Max.”

Behind the scenes, a data tree mirroring your train of thought and memories is now created. Next, inluu will ask you about the relevant topics you mentioned. In the above case, the training tool would drill down and ask you about subjects like Seattle, Dana and Max. And so it goes on, building a hierarchy of your memories, keeping topics connected to each other the same way they’re linked in your mind.

The more you share, the more questions inluu will have for you. The training process can go on forever; it’s up to you. You can spend hours in one session answering questions, or just ten minutes answering a couple, then return a week later. You can build your inluu in one sitting or slowly over time. It’s all up to you. Your answers can be public or password protected, and you can attach images to your responses, adding richness to your story. If you’re not sure what to write about, inluu will guide you right along. You’ll be surprised at how fast your inluu memory tree grows, each leaf representing a story of yours worth saving and sharing.

But what can your inluu do once it knows all this stuff about you? This question leads us to…

Chat

This is where your family, friends or the world (if you wish) go to chat with your inluu. Yes, chat. When people first visit, your inluu will introduce itself much in the same way you did during the training process. In your words. Then your visitors will be able to type natural language questions and your inluu will respond as best it can, based on the information it has about you. For example, my future grandchild might some day log in to my inluu and ask: “Grandpa, why did you have a Spanish name?” My inluu might respond:

“Well, I was born in Spain. Both my parents were Swedish but I guess they wanted to torture me by giving me something to explain to everyone.”

…along with a picture of my parents in Spain, laughing at their Swedish baby with a Spanish name.

inluu behaves like a bot, but it’s different. It does not make up its own stale, computer-like sentences. It does not sound like a robot. It sounds like you. All responses are in your exact words, the way you intended. That’s the only way it can be your inluu. It understands exactly how your memories and stories are connected because it interviewed you. It is able to stay on topic throughout a chat. The responses it comes up with to people’s questions are usually frighteningly accurate. Your inluu is pretty smart. It handles spelling mistakes, synonyms and related topics just fine, making chatting fun.

In addition to chatting, there are other ways to interact with people’s inluus. For example, when a response is displayed, keywords which contain further information are hyperlinks. This works much like a wiki; all of these links are automatically generated for the inluu owner. Using inluu.com is easy.

Your inluu won’t ever replace you, but hopefully it will help you leave something behind that’s worth more than money. Think of it as life insurance for your mind.

Love it or hate it, please leave a comment with your thoughts. Guy Kawasaki just made an excellent post on the virtue of ignoring, so I can take it.

inluu.com — Your story. Your people. Your pace.

A few people have contacted me and made it clear that I’m not making enough noise lately. Well, inluu is coming along, albeit at a slightly slower pace right now. The business lead, project manager, technical lead and developer all have had some pressing personal issues to deal with, all at the same time. (All their names also happen to be the same and they all look the same — it’s a weird office, I tell you.)

So, on a personal note, we have acquired a condo in downtown Kirkland and we’re very excited about it! Of course, we did the entire transaction through Redfin. This meant a high percentage savings on the commission (not insignificant with Seattle/Eastside real estate prices as high as they are). It looks like we’ll be moving this month and we can’t wait to spend the summer (and many future ones) within walking distance of the market, dozens of restaurants (try Thin Pan right on Lake St. for the best Thai food in Seattle), coffee shops (my favorite being Kahili), Performance Center, parks, water (with the Island Sailing Club) and most importantly events like the Kirkland Concours d’Elegance. This annual concours has quickly earned its place among the very best in the entire country; in my opinion second only to Pebble Beach. In fact, both Edward Herrmann and Keith Martin are on the Kirkland Concours team. Both, of course, have a long history of involvement with events like Pebble Beach and Concorso Italiano. (Here are my pictures from last year’s Kirkland event.)

So come on down, spend all your money and help drive real estate prices to new ridiculous heights! In between sorting through all of our stuff, packing boxes and imbibing countless cups of coffee, I promise to get inluu closer to beta testing.

/Juan

Two decades ago, my good friend Jim Malmros (of IO Interactive fame) and I had frequent discussions about the possibility of transferring, copying or otherwise messing with the contents of the human mind. Partly because we were weird but also because most computer science types probably wonder at one point or another just how similar the human brain is to a computer. Considering our maturity level at the time, I’m happy to report that these were purely theoretical exercises.

Since that time, the idea of leaving my stories and experiences intact for others to enjoy has fascinated me. I started thinking that it would really suck to have all the information stored in my brain simply vanish upon my expiration (there are those who would argue there isn’t all that much information in my brain to lose). But simply writing a book about myself wasn’t going to cut it. I wanted something people could interact with. Something they would come back to time and time again when they felt like a chat. Something that would surprise them once in a while, and something that would be available to anyone at any time.

If I was ever going to do something useful with technology, this was it.

When my grandfather passed away a while back without the family, despite our best intentions, having captured everything we wanted to, I knew I had to act. I spent a ton of time doing manual proofs-of-concept, and later implemented successful software prototypes. And here I am now, working on it double-full-time.

While a cure for death is a long way off (and I’m fairly certain I won’t be the one finding it), I’ll be happy if inluu.com can encourage and help people in saving and sharing their life stories. Far too much important information is lost when the people around us depart. Taking the time and spending the energy to write a formal autobiography, let alone getting it published, distributed and read, is a challenge only a fraction of the population is up to. Blogging is great but there’s nothing guiding you through the process, keeping track of where you are and what remains to be said, structuring your memories to mirror your mind. And the end result is not interactive. We need something that opens up the opportunity for endless but easy, fun data mining.

The mission of Human Proxy is to invent and implement technologies that help you be there for your people. The upcoming launch of inluu.com v1.0 will be the first, tiny baby step. The list of future features is long, because I believe the need is great.

/Juan

I bring good news.

Last Friday (or to be perfectly correct, Saturday morning) we hit a major development milestone. The core functionality of the Human Proxy engine — which powers inluu.com — became operational! At the risk of making it sound like that’s as cool as Skynet becoming self-aware, I’m writing this post to share my excitement with you. (Fear not, Cyberdyne was not involved in the development of this product.)

It’s one thing to have a vision for a product and dream about how great it’s going to be. That dream carries you through the rough times. But to see your vision come to life, the various components playing nicely with each other and doing what you intended… Well, it’s absolutely exhilarating! Last weekend I had the opportunity to do just that. Test it out. See it do things the prototypes could only approximate, thus proving for the first time that this is truly viable and is really going to work. That’s good validation after six months of intense work.

Special thanks to my wife Dana, who also comes from a computer science background, for putting up with endless white board sessions, the occasional creative language and sudden screams of both fear and joy. Also thanks to my ten-year-old son Max who has proven to be quite the software tester. He is the reason I had to spend an extra two days on the parser. I need to put in a plug for him here: last week he created his first HTML/JavaScript page from scratch. He’s comfortable with creating functions and passing parameters now. He thinks variables are pretty cool, especially the fact that you can divide 10 by bob. Next up is flow control. Poor child, you say? Well, he never had a chance in this family.

The next step is fine-tuning and optimization of the two engine components. I’ll describe what they do soon enough. After making the entire engine production-ready, there are a number of small, but not technically challenging, things to wrap up. Then we go to beta testing! If you’re interested in participating and you’re not already signed up, let me know.

After beta testing, there will be a pilot phase to build critical mass. I would love for you to participate in that as well! Indeed, the success of this venture may depend on it. Obviously, there’s a free life-time account in it for you.

Thanks for all your support!

/Juan

John Cook of the Seattle Post-Intelligencer has updated his great Seattle Web 2.0 list. I’d like to thank him for including Human Proxy!

John aptly described the company as “online memoirs”. I have only described inluu.com in detail to a select few people, and I tend to spend 30 minutes talking their ears off because I’m so excited about it. So far, the feedback has been very positive. John’s concise description is correct and I can’t wait to share with everyone the unique approach inluu.com takes to facilitate memoir writing and sharing.

inluu.com is not your father’s blogging tool — but hopefully it will be.

/Juan

This post will be shamelessly self-therapeutic.

I’m sitting here working on a particularly complex piece of engineering. It’s without a doubt the most intricate search algorithm I have ever had the pleasure of inventing and implementing. inluu.com will have two core pieces of functionality; they’re like two sides of the same coin. One of them is production-ready. For the other one, I need to complete said search algorithm and it needs to be exceptionally smart. (In essence, it needs to search like a human, not a machine. Less Vader, more Anakin.)

I’ve spent countless hours inside my own head (an interesting place) and at the whiteboard. Looking at the source code fragments is frightening and confusing yet oddly compelling. It’s like the partially completed Death Star in Return of the Jedi. But when I’m feeling stuck in the coding, being stuck in the tractor beam of the beast is what saves me. I know I’ll get past this hurdle because if I can’t figure this piece out, there will be no inluu.com. And I’m committed to inluu.com.

Which brings me to something that hit me a short while back: I’m so glad I committed to this venture before I thought too much about the deep details. My initial proof-of-concept turned out great and that’s all I needed. How many people have ideas they never follow through on because of the what-ifs? I’ve seen many people’s lightbulbs instantly dimmed by huge, stormy analysis clouds. These people’s minimal investment in the idea is easily outweighed by the challenges they immediately see before them. They don’t take the time up front to commit to the journey before falling victim to their own weather forecast.

I say “take the time to commit” because I don’t believe one can typically commit to something in a matter of minutes. Taking the time to commit entails working up a true belief in your idea to the point where you are completely (read: fanatically) energized about the whole thing. You must be able to see before your eyes how the finished product is already helping people and how they have giant smiles on their faces. How they can do things they weren’t able to before. How you built something great. How you’re changing at least a small part of the world and how you and others are benefiting from it.

I’m not saying you should ignore business- or market realities. Unless you’re a non-profit, the idea itself must be commercially viable or it will be difficult to continue helping people with your product. What I am saying is that once you think you have something, spend some quality time committing to it, keep the bad clouds at bay and then go for it! Then, when the clouds eventually do roll in, they’re just weather, not instant idea annihilators. And that follow-through and persistence — that commitment — is really what will give you a real shot at success, isn’t it?

for ($i = 0; $i < $syn…… oops, sorry, wrong window.