Rev2.org

When I first saw the demo of FireFly, my instant reaction was, “no freaking way!”. Turns out I was wrong, and yes, there is one freaking way — it’s called FireFly. Launching recently at the Tech Meetup in New York, it’s a product of combinator/VC firm BetaWorks. Undoubtedly, it’s received a response no less than amazement in the blogosphere (Dave Winer wrote about it as did Allen Stern.)

The best way to describe it is this: it’s a simple flash widget that anyone can embed which lets you, and your site visitors, watch and interact with each other at a given moment. Obviously, this means visitors can chat to each other, but here’s the cool part: they can practically watch the mouse movements of other visitors on the site at a given time. What they’re clicking on, what they’re highlighting, everything.

It’s a form of spying, I’ll admit, but it is ridiculously and very freakishly cool. The site it is currently in a private beta, and I was unable to send a request by the time I wanted to get this post out, but you can see a working demo on their site — just click the “View Demo” button. If the FireFly folks are reading this — I would love to get an invite so I can embed the widget to this post.

Why would such a technology be useful? I have no idea. Visitor interaction or something like that. But the point is, it’s possible. And there definitely a lot of useful applications. To put this into context, it is basically using a simple web browser and Flash to multi-screenwatch a number of people’s mouse movements. Now that is something.

The book binded together a lot of information into a portable format. The eBook questioned it — asking why it had to be something physical and with a hard cover when the digital age allowed for information to be so much more. The eBook Reader, or Kindle, took that forward, combining the portability of the book and the ubiquity of the eBook. And now, DailyLit is taking a horizontal step toward a new concept — subscription-based reading via e-mail or RSS.

Here’s how it works. Like any book store, DailyLit has a bunch of books for everyone to buy. Most free, some a nominal $6.95. You can search through their catalogue, or browse through categories. When you’ve made the purchase, you’re not going to get a hardcover in the mailbox a next day or something to download, but an RSS feed or e-mails going/updating periodically.

Yes, it’s a weird concept, and definitely one that might need some getting used to, but for a daily commuter with an iPhone or BlackBerry, an RSS fanatic, or an e-mail-head, it could  prove as somewhat of a solution. And the best thing is, since most books are free and in the public domain, you don’t have to buy and try — just pick a book and see if it suits you.

Back in 1999 when you were working on your Angelfire homepage and waiting for your My Excite homepage to load, if I were to tell you the idea behind this service, you’d laugh in absolute ridicule and question the need for it to be a web service, let alone exist for the everyday user. Today, it’s possible and it works. It’s FontStruct, a new service that lets you create and share your own fonts (literally) using its Flash-based builder tool.

Font builder tools have of course, and obviously, been around since the personal computer has existed, but this is the first time I have seen the idea ported over to the web. The way it works is simple — signup, name your font, and go through each letter to build it using their tools.

The service gives you a number of ordinary and useful tools such as a pencil, eraser, rectangle, and select, but where it shines is with its font-specific options. This includes one for diagonal lines, and various other ‘bricks’ for rounded corners, circles and such.

Once you are done with designing the font, there are several things you can do with it. You can preview it, save it, download it as a standard TrueType font, clone it, embed it into web space elsewhere, or delete it. You can also change its sharing features to make it publicly accessible and listed in their very impressive gallery.

I’ve only ever had an interest in typography from the designer-consumer aspect and never from actually designing the actual font, as I’m sure would be the case with most Fontstruct users, and the tools they present are helpful and give a good background when it comes to making your own font. Moreover, to the average 1999 Internet user, it’d probably be clear as to why the service is web based when it could be a simple .NET app: sharing, community, collaboration. These are elements the typeface industry always carried, and FontStruct just brings it all together.

Update: Thanks to the Tunesbag folks, we have 30 invites for Rev2 readers! Follow this link and sign up — the first thirty will be able to get in.

Tunesbag is a site I was pointed to by a friend, and recently let into to their invite-only beta (request to the Tunesbag folks: Rev2 readers are craving invites). The best way I can describe the service is like an iTunes for the cloud — similar to a number of music sites we’ve seen (namely Muxtape comes to mind) but much, much more open and feature-rich.

The service is complex in that it has a number of features geared toward helping the user, but a user who has used iTunes should grasp the interface fairly quickly. After logging into the site, you’re taken directly to the dashboard page. Here you can upload songs in various ways, invite friends and start sharing your music, or switch to the library mode where you can play the songs.

Uploading
There are a number of ways to upload to Tunesbag, certainly useful to the user and feature-intensity put in the right area. You can download their native software for use with iTunes, Windows Media Player or Winamp and upload directly, you can upload manually using their multi-file select Window, your can point directly to the MP3 file on the web, or a personal favourite, e-mail your MP3(s) to username@incoming.tunesbag.com

Socializing
With a Facebook app and some integration, Tunesbag allows you to make use of your social graph at Facebook to share music with and get music from them. Tunesbag allows you to invite your Facebook friends directly from the dashboard, and once they’re there, they are able to access your music and vice-versa.

Playing
If you’ve moved around computers and have had to transfer music, or worse still, lose them, you’d know the value and having it backed up elsewhere, especially on the cloud. Alternatively, if you’ve been away from your music and had the urge to access it (I know I have in the pre-iPod days), it’s not a good feeling — and it begs the question that if you can store your e-mail and documents and files in the cloud, then why not music? Tunesbag’ library solves this in that it’s the iTunes/MP3 player for the web. You can play your music, manage the meta data, or try different modes.

Conclusion
Piracy and legal issues aside, Tunesbag is a great service. I love the idea of having my music elsewhere and not relying on iPod/iTunes for the life of them, and certainly the social sharing application there is useful and going places. That said, I do hope Tunesbag is able to avoid the Napster situation and stay out of legal troubles. Sites like Muxtape have succeeded so far, but Tunesbag has a much open model — and specifically encourages the concept of sharing. If they’re able to stay out of the record industry’s radar, they’re going to get somewhere.

“Finally, something!” are the cheers in Silicon Valley this morning. Powerset, the mega-hyped underdog Google killer semantic play, has finally launched a working demo of its product. While the search engine has been in closed beta for some quite now, this is the first time it has unveiled any of its technology to the public.

What is being unveiled is essentially a search engine based on top of Wikipedia. Allowing you to enter a topic, phrase or question, Powerset uses its semantic algorithm to dig through the keywords and come up with the best matches — or better still, answers — based on the Wikipedia articles. While this is in no way a final product, the demo is supposed to show the glimpse of its algorithm on a test index.

Trying to differentiate from their suggested examples, I tried a few queries, and I am no less than impressed. The first query I tried was “who is Andy Kaufman.” Rightly enough, Powerset was able to extract a mini-bio from Wikipedia, structurally providing a brief description, date of birth, date of date, profession, film roles, etc. What I was more impressed was their ‘factz’ section where key facts are analyzed and put into context. While not totally accurate, I was able to find out names of the characters he portrayed among other things.

The second query I tried was, simply, “Iron Man.” To my surprise, a special character/name supplement was presented again, and this time with more detail. Among the tabs included “Iron Man,” “Film,” “Video Game,” “Song,” “Magazine,” “TV Series,” and “Band,” presumably all taken from Wikipedia’s disambiguations. Going through, I was able to get a good idea of each, although in brief snippets.

For my final query, I decided to try something a little more obscure. This time, my query was “who invented the computer mouse?“. Rightly enough, the first result was Douglas Engelbart’s Wikipedia page, with (in highlighted text) “He is best known for inventing the computer mouse”. All over the results, his name was repeated again and again, too, and tied closely with “invented” and “computer mouse,” so I could be sure it was him just by glancing.

I’ll admit, Powerset’s first display isn’t a bad one. It does the trick, and definitely defeats Wikipedia’s own search in a heartbeat (but then again, that’s not hard to do.) For the general queries, especially those which you’d expect to have a Wikipedia page, the results are more than average, O.K. for slightly obscure ones, and kind of clueless for the more obscure ones. It does not bad for a search engine that has been in stealth mode for 2 years, but certainly, we’ll need more to judge. I’m waiting for more the Google killer till it gets a pass for the hype its received. Oh, and I swear if I hear the word semantic again, I’ll …

In the midst of writing about newly launched hype-machines, I’m always flattered when I come across a tool that I haven’t heard about, but is well-built and already accumulating users. Such was Dipity, when I saw an embedded widget on the blog of a certain webebrity. The site aggregates services a la FriendFeed and other ‘activity aggregators’ (© Sid Yadav 2008) but presents them in a neat way.

The service functions on timelines — you can use the site to browse through the ones that have been created categorically, or create your own. At its core, it takes RSS feeds and puts them into a visual timeline — nothing technologically fancy, until you see the results.

To create one, you simply sign up and supply it with usernames to various online services, or any RSS feed. Currently, Dipity natively supports most of the usual ones like YouTube, Twitter, WordPress, Last.fm, Flickr and Yelp. Once you’ve inputted your presences, Dipity gives you a timeline — and that’s the end of it. Here’s mine.

The timeline is scrollable, and since (unlike others) Dipity indexes your past first and adds with your present, you get to see the results instantly. The Twitter posts are definitely the most telling. You can also zoom in to get a clear view. But although the service is mainly focused on timelines, that’s not the end of it. It provides you with a list view, flipbook (increbily cool), and a mapview for your geo-tagged items.

It’s amazing to see your online life represented in this way and definitely something — if you sign up for now — you’ll look back in a few years if you still remember the service (and it’s still around.) But of course, it’s not just for your own online life — the service works great as a timelining service in general, and there are a number of neat ones created. As a recently turned Iron Man-fanatic, this one caught my interest (and believe me, there are some other neat ones if you look).

Update: we have invites! Head on over to http://viewzi.com/invites and enter “gio” as your referral code.

Visual search engines, or rather, the ones that make a claim to provide alternative visual ways to search, have been on the rise lately — and rightly so, with the advent of Flash and AJAX. A few days ago we reviewed SearchMe, which featured an Apple-like ‘Coverflow’ view of its internally-crawled search results. Viewzi is the latest play in the area and is being touted in the blogosphere as the first to get it right.

The search engine, firstly, differs from the few I’ve seen in that it doesn’t assume that only one type of an interface is suitable for every search. Instead, it presents and allows you to choose from a slew of different interfaces and tools after you make your query. These are slightly optmized to the query, but I didn’t find much of a difference between the ones I tried — perhaps its because of the limited number of search tools available at this point.

There are, obviously, a whole bunch of tools to select from, but I’ll cover the main few that were presented for most general queries (note: unfortunately you need beta accounts to click on the ‘example’ links. If the Viewzi folks want to help out invite-craving Rev2 readers, we’d love some):

Simple Text View (example)
Loads simple/regular text results from meta-querying Google and Yahoo!. While fundamentally the same as normal search results on the respective search engines, they are slightly more edge and ‘2.0′ and sleek with their highlighting, and provide a helpful Compete.com rank for each site.

Web Screenshot View (example)
Similar to SearchMe, the results with this tool are presented in the Apple coverflow style, although not so mimetically. Additionally, since Viewzi is a meta search tool and doesn’t crawl its own index, the results are from Yahoo!, and while drastically more relevant than SearchMe’s, the screenshots are of inferior quality.

Basic Photo View (example)
This is the image search function of Viewzi. Searching Riya and Flickr, the tool gathers images and presents them in a more navigatable format. Although, it links back directly to the original image as opposed to dynamically enlarge it unlike most other visual image search tools I’ve seen, which I thought was a little odd — if that’s the point, why not just use Flickr?

4 Sources View (example)
One of the strikingly different tools out there — and in a good way — this is like meta search 2.0. It searches Google, Yahoo!, MSN, and Ask, and lays out screenshots from their results with corresponding labels as to which search engine it came from. For ones presented in multiple, or all 4, the screenshots are stacked. I’ll admit, it’s no less than neat.

Video x3 View (example)
One of the more nichely useful tools I found, this one searches through video search engines — namely YouTube, Blinx, and Veoh — and presents the results in a scrollable landscape video reel-like format. For a video searcher, it’s not a bad searching alternative as it is. I also love how the Blink search results are animated.

Everyday Shopping View (example)
This is one of my personal favorite tools. It searches shopping sites — Amazon, eBay, Walmart, and Target — and lays out the results in a simple, standard format with the picture and the price. For a shopping searcher, this could prove immensely helpful. And in the interface lickability factor, it doesn’t score too badly either.

Conclusion
After reading through the hype and testing out the tools individually, I have to admit, Viewzi lives up to it. Something I love is that as a search engine, it’s focus is not on relevancy or providing the best search results, but on the different ‘views’ and how best to present them using search algorithms already out there. And it works fantasically. I’ll admit, there are a few clunky tools — such as the Photo View and Screenshot one that I’m not sure add a whole lot of value — but there are some real jems.  And the best thing is, as more get built, it’s going to keep growing with more new tools. For a visual searcher, for the 2.0 searcher, and for the advanced searcher, Viewzi isn’t just worth a try — it’s worth the replacement.

UPDATE: First 8 comments get an invite!

Ed. Note: Apologies for the down time! I’ve been immensely busy between travelling, setting up a home office, and working on a startup. Hope things can get back to normal from here.

Mento, founded by Greg Hochmuth, is a new service based out of Germany and that entered private beta very recently. The service is based around the idea of sharing links with the people you care about. It’s not quite Digg, it’s not quite Tumblr, and it’s not quite Mahalo, but a weird mix of all three with a bit of its own.

The service is a easy signup — one of the more intuitive processes I’ve seen recently. At the end of it, Mento gives you a handy “Share with Mento” bookmarklet, which is a main input mechanism into the service. Whenever you’re visiting a link, pressing the bookmarklet reveals a bunch of actions that can be taken with the link. This also defines Mento pretty well in general — you can send the link (supplying a friend’s name/linked with Facebook, an e-mail, or a group/tag), save/bookmark it del.icio.us-style with a description and tags, or add a reply to a recently-bookmarked link on the site in your network.

While the input/sharing mechanism is one main aspect of the service, it’s not the only one. Going to to the site’s homepage, mento.info, reveals the “receiving” end of the spectrum. Among this is your own “Mento Mix” which shows the links shared by your friends and network recently, an “Inbox” which shows the links and replies sent specifically to you, and a way to customize your Mento Mix which is a way to tell Mento the kind of stuff you like to see and don’t like to see.

Additionally, the “Your Links” tab on the top shows your own recently shared links, “Your Network” allows you to manage your contact list and add/invite new friends, and “Channels” lets you join specific groups and categories of like-minded people with similar interests who collect and share links together. This aspect of the site is very much like a YouTube for links where the focus is on aggregation the whole community’s shared items as opposed to the links shared to you by your friends (which is the site’s main purpose).

A really cool thing about the site is being able to link and work with other services. For example, you can connect your Facebook account so that linked shared with Mento are automatically shared with Facebook as well. Additionally, you can export your links to FriendFeed, Twitter, Magnolia, Tumblr, or Del.icio.us all at once, and even use Twitter as a e-mail/bacn replacement to get notified about replies and recommendations.

Aesthetically and simply in the way it works, Mento is a solid product. There has been thought put into its design and interface and the way it looks, feels, and works very much serves to its purpose of building a community around link sharing. Being a heavy Tumblr, Del.cio.us, Twitter and FriendFeed user, I still have to try its integration points, but if it works the way it’s supposed to, I might as well use it as my primary bookmarklet to share something interesting on the web as opposed to twittering, tumbling, and del.icio.us’ing one link three times.

And as the site itself gets built outward and collects more users/my friends, I think I could find a use for it myself in discovering and sharing links. Personally, I think the service is on its way to being a success among other services like Twitter, Del.icio.us, and Tumblr in the area of link sharing if it’s able to somehow gain a good tipping point. But don’t forget, I called it first.

First, it was Dodgeball — the first supposed local social network. That got bought by Google and dumped into the trash can. Then it was Twitter. It’s evolved so much more that “what are you doing?” has lost its value. Recently, Yahoo! launched its own play FireEagle in private beta (we covered it here) which is starting the location-craze again — with plans to have multiple input/output methods looming across the web and devices. But that’s not so much of a location-based social network as it is a “where are you, exactly?” version of Twitter. Here’s one that is — BrightKite.

Currently in a private beta also (first three comments get an invite!), the service is based around locations and what you and other people — your friends, the world, etc. — are doing at what places. After signing up, users can search for their location in the places box and “check in” — similar to Fire Eagle, a Facebook status or a Twitter message. After checking into that place, friends are updated, and you can see who is within your proximity or where your friends are at the moment in the same fashion.

Once the user is checked into a place, there are a number of actions that can be taken. On that place itself, a 140-character Twitter-like note (i.e. we went to Alcatraz today!) or a photo can be posted, and recorded in almost the form of a travel diary — except that it’s ongoing. An additional step that you can take is opt into their mobile service and “placemark” your current place — useful if it’s a specific location. So, for example, where you’re at work, you can check in by texting “@work” to 80289. This gives identity to places and lets you easily let your friends know where you are.

Apart from travellers, whom it’s obviously extremely useful for, I think a huge potential market is urban hipsters. With Twitter being turned into a more of a “broadcasting” mechanism than meeting up and seeing what your real friends are upto, not to mention the vast amount of noise and busyness that fills the streams, BrightKite has the potential to be an invaluable tool.

If there ever would be an eBay 2.0 (or an eBay for the Web 2.0 generation), here it is. Wigix, which stands for “WantItGotItExchange,” is a newly launched service based around the idea of a community-driven marketplace. The site takes the idea of eBay — a community-powered auctions market — a step further, by adding a stock market-esque element of being able to “buy” and “sell” the product at a price determined by the market.

Here’s how it works. Like eBay or any other community shopping market sites, users can buy and sell products. The kicker is: the price, rather than being determined by the seller itself, is determined by a standardized market at Wigix. The service lists market products and has standardized pages for each (saving time and avoiding the seller having to create it for each listing). Then, following the concepts of demand and supply, the price is determined by the number of buyers or the number of sellers a product has. The more the people buy a product, the lower the price gets, and vice-versa.

So in concept, the market decides an equillibrium price for both the buyer and seller, and depending on the particular situation and the price of the product elsewhere, they can decide whether the offer from Wigix is worthy enough. Like most stock markets, the price will more often than not be beneficial to both parties with the equation of time in place (should I sell now or anticipate a higher price and sell later?), which is what makes it work.

The idea, in concept, is needless to say nothing less than brilliant. There’s a value proposition for everybody: sellers don’t have to spend the time and effort creating product listings and deciding prices, and buyers don’t have to look through 10,000 listings and decide which one to bid on and scan through seller ratings. Additionally, the service boasts no listing fees and no fees for items under $25. If you’re looking for an alternative to eBay or Amazon, Wigix is a definite contender.