Saturday, June 02, 2007

RIP Steve Gilliard

I never met Steve Gilliard, but I used to read his blog all the time, so I sort of felt like I knew him. He died today, too young, at 41. He'll be missed...

Friday, June 01, 2007

Something's Brewing in this Town...

Okay, so, I've told you all to watch Windy City Heat on the blog before. It's the amazingly hysterical prank movie in which several comedians band together to convince an idiot named Perry that he's starring in a new action movie. Of course, there is no movie; the whole thing's a set-up to humiliate this guy, Perry Caravello, who dreams of stardom despite being an unattractive, paunchy middle-aged guy with no discernable talent.

Windy City Heat is one of the funniest films of our present decade. I am not even kidding. A work of fiction could never tell such a ridiculous story - it's nearly impossible to believe that anyone could be this dumb. In fact, many people to whom I have shown the film have insisted to me that it must be fake. Perry falls for such obvious set-ups - bad Charlton Heston impersonators, Japanese investors named Hiroshima Nagasaki, an action movie script with obviously jokey dialogue and no action scenes directed by Bobcat freaking Goldthwait - that it almost has to be an act.

Well...maybe not...From TMZ today:

Perry Caravello, the one-time "star" of a practical joke flick, filed suit today in Los Angeles County Superior Court claiming, among other things, the three funnymen owe him $10.5 million.

For several years, Caravello was duped into believing he was going to play the lead role in the faux action flick "Windy City Heat." Every actor and member of the crew was in on the joke -- except Caravello. The final product, er, joke, aired in 2003 on Comedy Central.

In his suit, Caravello claims the defendants "falsely and fraudulently" represented the project, and promised him, among other things, that he "would be paid 10 million dollars if he placed his penis in a mousetrap." Caravello says he "was severely injured when the trap literally went on his manhood." Ouch. To add insult to injury, Caravello says he's suffered "humiliation and emotional trauma" because the video has circulated the Internet. Talk about viral.

Caravello is suing for the $10.5 million that he says he is owed, plus damages and medical expenses.


First of all, let me just say that I'm deeply upset there was a video of Perry Caravello placing his penis in a mousetrap online and I was uninformed. Now, I've been really looking for a little while now, and haven't turned up anything (perhaps I'll have more success with Bit Torrent), so it's entirely possible that Perry's full of crap on this one and looking for a payday.

But if not...then I think this confirms that he really is stupid enough to fall for anything, up to and including a fake movie called Windy City Heatin which he'd portray Chicago's greatest sports detective Stone Fury.

Thursday, May 31, 2007

Yes, Idea Man! Creator! Innovator! Cerebrator! Tycoon!

Yes, the Idea Man! What're his hopes and dreams, his desires and aspirations? Does he think all the time or does he set aside a certain portion of the day? How tall is he and what's his shoe size? Where does he sleep and what does he eat for breakfast? Does he put jam on his toast or doesn't he put jam on his toast, and if not why not and since when?
-- The Hudsucker Proxy, Ethan Coen, Joel Coen, and Sam Raimi

Back in January, I answered a mysterious Craig's List ad for researchers. No information on what would be researched. No office to report to. No company name provided. I responded anyway, if only to find out what the hell was going on.

A few days later, I found myself walking around a stranger's home in Brentwood. I had been told to appear at this address, but there were no signs, no instructions, no people waiting around to guide my way. I walked through a gate and into a backyard. In a guest house next to a pool, a few people were busily working on brand new iMacs, but they were consumed in their work and not looking my way. No one else was around. The thought occurred to me that I should leave, that this could be a scam. An inscrutable scam, one that involved fraudulently luring in job applications, but a scam all the same. ("Okay, where's Ashton? Oh, right, I'm not famous...")

It was not a scam. I soon thereafter met Mark Jeffrey and he pitched to me the basics of Mahalo, what was to be the latest venture from entrepreneur Jason Calacanis. Two thoughts crossed my mind simultaneously:

(1) This sounds like a fun and interesting job!

and

(2) This idea will probably not work!

I assumed, quite naturally, that because the Internet is composed of millions of pages about millions of topics, any one of which could be of interest to anyone at any time, there would be no reasonable way for human beings to neatly organize all available information. This is the reasonable conclusion to reach. Anyone who has spent any amount of time online would understand this point.

After 5 months on the project, I am now quite confident that this is not a major concern. Not because Mahalo will be able to respond to any potential query anyone could have, but because the actual range of commonly-searched queries is quite manageable for a large enough staff of Internet "guides."

Let's take a look at this Donald Trump page. Now, people talk about Donald Trump all the time, everyday, all of the Intra-t00bz. That's his job. He says or does pretty much anything he can think of to ensure that people will talk about him. If I personally have hundreds of Mahalo pages to update, create and oversee, there's no way I could possibly find the time to put a new link on this Donald Trump page every time he insults Rosie O'Donnell. Honestly, even if I was only in charge of updating the Donald Trump page for 8 hours a day, I still wouldn't manage to get all of his Rosie insults on there. The guy's a machine.

If you'll skip down to the Gossip and Blogs section, you'll notice that we have syndicated the 3 most recent Gawker posts. Doing this in both the Gossip and News sections of the page make it, essentially, self-updating.

Naturally, someone will have to check in with the page now and again, to clear out dead links and find new articles and videos of special significance. Say, if Trump gets a new TV show, or starts picking on a different daytime television host. (Might I recommend Rachael Ray?) But even if we don't get to it for a week or two at a time, new news and information from quality sites without spam will always be available on Mahalo, 24/7. (Not to mention the fact that users can suggest their own links, which get immediately e-mailed to the guide who created the page, allowing us to update and incorporate new links at any time quickly.)

I didn't just want to brag about my involvement with Jason's remarkable invention; I'm actually building to a point here. See, the idea itself for Mahalo, no offense meant to Jason...it's not that hard to come up with. You don't have to be some crazy futurist visionary to imagine a version of Google that's compiled by people, so it doesn't suck quite so bad.

A lot of search engines have even come fairly close to the same idea. This one pairs you with a live guide in a chat room, who does your search for you in real time. (It doesn't really work all that well.) This one clusters your searches for you, making them more organized and easier to scan than many Google results. This one organizes sets of sites for you before you search, allowing for more exact, targeted results. (It does appear to limit the number of sources you can search, though, instead of giving you a taste of everything that's out there.) And the guy behind Wikipedia is working on a somewhat similar concept as well, sort of a Search/Wikipedia hybrid allowing individuals to create website-portals about any topic.

But anyway, my point is, what made Jason's Mahalo concept so special wasn't the idea itself but that he's insane enough to actually attempt to make it work. Had I come up with the idea of human-powered search, I would dismiss it out of hand immediately. Couldn't be done. Too many searches. It would take too many people. You'd have to outsource the work and would get crummy results. Spammers would get through.

Of course, Team Mahalo has still found a way to make it work. It wasn't always easy, and not everything we all theoretically want to do with the site is possible, but you can go check it out for yourself. We're off to an amazing start and it's only going to keep getting better as we go.

People have this very strange idea about ideas, as if brilliant innovations just shoot out of the foreheads of geniuses, Athena-style, who and complete. It's nonsense. The great breakthroughs are all assembled slowly over time, building on mistakes and improving gradually, because solving problems is much easier if you're willing to leave them unsolved for a while.

Is this obvious to everyone else, and has only taken me 28 years to figure out?

Because my instinct is the exact polar 180 degree opposite of this philosophy. When I encounter a problem - any kind of problem, really, personal or professional - I want to solve it immediately, right away, or I get all antsy and upset. C.K. has actually called me on this pretty recently; I meet with some difficulty, and I can't stop obsessing about it until it's taken care of, one way or another. This is, to put it simply, not how things work in the real world. Sometimes, things don't work quite right, and you take the time to figure it out, and then you fix it.

When we were first building these Mahalo pages, we had no idea what we were doing. There was no template for a human-designed search results page (or SeRP, to get all insider-y on you). A few of us guides invented it, and then scrapped that design and went with a new one, and then scrapped it again. We had all sorts of ideas that didn't come to fruition: massive cartoon graphics charting celebrity hookups, interactive maps of the "Lost" island and on and on and on. It's been an enlightening experience to make mistakes and keep making them, knowing that you'll figure them out later. It's...I don't know...relaxing.

Anyway, there's a major life lesson in all this somewhere, but damned if I can see my way through to it, exactly. Something about having the courage to actually believe that something might work, that one idea might actually be worth trying. It's not arrogance, exactly, although it can resemble arrogance. More like a particularly aggressive strain of confidence. And I can teach you all about it for the low low low price of only $39.95. Call today.

Tuesday, May 29, 2007

Mahalo

As attentive readers will recall, this past January, I quit my old job clerking at the Laser Blazer and started a new gig, referred to here as the Shadowy and Mysterious Project Which Can Not Be Named (or SMPWCNBN, for short).

Well, today, I can finally reveal my Forbidden Job of Mystery. I have been working at a guide for the brand new search engine known as Mahalo, which you can check out for your very own self RIGHT NOW here at Mahalo.com. I encourage you to do so at your earliest convenience. If you'd like to see all the pages I have worked on since January, look here.

When I joined in January, I was Guide #3 (the less fortunate Guide #4 entered the room an eventful 12 seconds after me). As of now, I'm working on a team of about 30 guides, with an additional crew of developers, podcast engineers and other various and sundry online Internet-ish types. It has been quite a ride, one that I've wanted to tell you all about, Dear Readers, for months now.

But we were operating in stealthy mode, so I was not at liberty to reveal key details...like meeting PayPal founder and Mahalo investor Elon Musk, pitching the site to bigwigs from NewsCorp. and CAA, winning a $500 Apple Store gift card, which I promptly used to obtain a kickass new 20" iMac with all the trimmings (and, yes, I could a Video iPod as "trimming") or getting my picture on the front page of Valleywag, where some commenter immediately noted my physical similarity to a certain overweight bearded film director.

It has been an eventful couple of months to say the least. And now, after a lot of hard work, free lunches, triple espressos and long conversations on the nature and identifying characteristics of splogs, we're ready to launch a brand-new search engine.

BUT NOT JUST ANY SEARCH ENGINE!

Seriously, Mahalo is going to change the way you all think about search. I swear. Honest.

For those of you who don't already know this, your popular search engines...your Googles, your Yahoos, and to a lesser extent, your Asks, they all function via the same mechanism. You type in a term and a sophisticated computer program sifts through the many millions of web pages out there looking for keyword matches.

Now, some of these programs are extremely clever. In fact, "semantic search" is fast becoming all the rage, which allows computers to read the way humans do, scanning collections of words for meaning rather than strictly keeping to keywords. Over time, Google's results have become more and more accurate, and they still do pretty well if you're looking for pictures of a celebrity or movie times or word definitions.

But, necessarily, humans will always be better at this than machines. I can use my creativity and my knowledge of a topic to deduce what another person would be looking for with more accuracy AND flair than any machine. If you type in "large hog" on the Internet today, all Google knows is that you're looking for a pig and you're looking for a page with the word "large" on it. Maybe it'll be a bit more clever and look for pages which appear to use the word "large" to describe a "hog," but that's a big maybe.

Also, you'll get lots of spam. Cumbersome-porcine-animal-photos.blogspot.com will probably come up. Maybe a few sites that look like they're going to be good but then turn out to have one or two outdated bits of information amidst a sea of Google ads. A bunch of sites that aren't what you're looking for - "Review: Wild Hogs Largely a Waste of Everybody's Valuable Time." Then maybe, maybe, if enough people are also looking for it, a link to the photo you really wanted.

But I'd know right from the start that this is really what you're looking for.

Now, imagine that for every search you'll do all week on the Internet, if there were someone on the other end actively trying to figure out what you wanted and finding it for you without all the legwork and irritation. That's Mahalo - human-powered search. Because humans are better.

Have you seen these Ask.com billboards that talk about "the algorithm"?



They're not kidding. All their results are based on algorithms; very smart computer programs. But, you know...not too smart...

All Mahalo results are based on myself or one of my co-workers ACTUALLY SEARCHING THE INTERNET and finding the things you want for you.

This is a powerful idea when you start to consider the ramifications. If Wikipedia is a way to catalog all of the world's information, this is really a way to curate all the world's information. I don't need to tell you that the Internet has an ridiculously massive amount of stuff on it, but no one is getting the most out of it, because 3/4 of total Internet time is taken up with getting to where you want to go. In the time it takes you to search something on Google, and IN LESS TIME than it takes to search for something on Digg or Del.icio.us, Mahalo can pretty much get you where you want to go.

Okay, so that's the sales pitch. The reality is, just as I said, this is a ridiculously massive job. Mahalo as of now has a bit over 4,000 pages. When we hit 10,000, we'll enter Beta mode. When we get to 25,000, we're officially ready for action. You'd be surprised how far 4,000 pages will get you. We already have every major sports team, most notable politicians, tons of fashion labels, products and gadgets, a bevy of film stars, TV stars, musicians, shows, movies and bands. My brother, who's also on the job, managed an entire section of Food pages, from Brie Cheese to Jack in the Box. (Check out the Tacos page, which helped introduce me to my new neighborhood haunt, Don Felix.) Most hot vacation destinations are done; most popular car models already have pages.

It's not terribly hard to "beat the system." Perhaps the most popular search that leads people to this blog, the centrally-important "albino porn," does not yet possess its own Mahalo page. I'll get to it, maybe next week. But you get the next best thing - Google results - and the chances are, if people start searching for something, we'll get to it soon enough.

I just realized that I've babbled on at you at length already, and haven't yet discussed any of the other amazing, totally blown-out Mahalo features. So I will quickly run down more about teh new search hotness:

  • Pages have Fast Facts in case you just needed to know something quickly, like the number of people who listen to Rush Limbaugh each week (20 million) or what day we celebrate National Corn Dog Day (the first Saturday of March Madness)
  • Searches that have not yet been completed can be requested, and we'll e-mail you back when they're done
  • Users can submit their own recommendations for pages we might have missed
  • Mahalo pages are easy to e-mail to friends
  • Mahalo pages often feature useful add-ons like videos, RSS feeds and stock charts right there on the page
Obviously, after putting in so much work on this thing, I'm anxious to see how people will react. But having said that, I'm fairly confident in the quality of Mahalo itself. The thing is just faster and more useful than any search engine, provided you get a result for your search. (And 10,000 search terms accounts for a full 25% of all Internet searches, so your chances of success are pretty good, even on the first try.) I'll be saying much more about all of this in the days and weeks to come (as well as possibly working on another blog in addition to CBI, all about the world of search), so stay tuned for more. But we're launching even as I post this at the Wall Street Journal's D: All Things Digital conference, so I at least wanted to get the word out. Enjoy.