Reviewables, the Official Notches Blog

Changes at Notches

To this point, we focused on providing the base infrastructure and REST API for others to build on. Unfortunately, this excluded potential partners that did not have the technical resources and know-how to integrate the platform. While the REST API will certainly remain an option, going forward we will focus on delivering tools that simplify integration for partners. Our goal is to enable any partner, whether a community or e-commerce site, to add reviews and product data to their site in seconds.

As mentioned on Mashable, we have built a Reviews application for Ning’s OpenSocial Application Directory which will allow community owners to add basic reviews functionality with just a click. I can't go into too much detail until the public launch, but we've gotten some great feedback during the ongoing private beta and we’re excited to release this to a wider audience. There is a lot of untapped expertise in over 1 million Ning communities and we’re excited to help unlock this and bring it to the Notches network.

We are also launching Notches Wire into private alpha . Wire is what we call "Disqus for Reviews" - a plug-in that allows e-commerce and other partners to easily integrate reviews functionality. We're still looking for a few more partners to help us test before we launch this publicly, so if you're interested request an invite code.

As we enter a new phase of the business, we are also undergoing some management changes. After discussing with Corey, we agreed that new leadership was necessary to lead us through this period; as such, he has stepped down as CEO. While no longer involved in day-to-day execution, he will remain involved as an evangelist and advisor. I want to thank Corey for everything he's done to get Notches to where we are today and wish him the best of luck as he moves to a full-time role with URDB.

Notches is presenting at the NY Tech Meetup Showcase

We will be presenting at the NY Tech Meetup Showcase tomorrow (Tues, June 2nd) from 3:30-6:30 at the Fashion Institute of Technology’s Great Hall. The showcase is free and open to the public, so we hope you will stop by.

If you are interested in being an alpha tester for Notches Wire, a tool for easily integrating reviews into your e-commerce site, stop by and we’ll give you a brief overview and hook you up with an invite code.

The Barnacle and The Boat

Back at the nextNY October Community Conversation, David Kidder talked about the differences between being a satellite company and a planet company. We use the barnacle and the boat analogy, but the point is the same – there are companies that stand alone and have gravity, and there are those that exist by attaching themselves to other companies that have that gravity.

Obviously, it takes a lot longer to build your own boat than it does to attach yourself to someone else’s. As a result, the boat is ultimately worth more and more resilient – something that, as David stressed, is increasingly important in this kind of economic environment.

In other words, building on the Facebook platform is a great way to gain “easy” access to a lot of users and traffic, but it’s important to remember that ultimately Facebook “owns” those users. If Facebook blocks your application or launches another redesign which further downplays your visibility, will your business survive? Can your business stand alone outside of the traffic you take for granted?

The common element between Facebook and other “planets” like Google and Microsoft is that they’re building core value that stands alone and, where applicable, creating an ecosystem around that. Facebook has this influence because of the number of users it has, not technology – as it demonstrated by recently open-sourcing much of their platform.

For those of us trying to build new, sustainable businesses, I think it comes down to finding the right balance between being a boat and being a barnacle. If you focus solely on the long-term value, you will struggle with traction and attracting users. If you focus on solely on leveraging users that exist on other platforms without "converting" them along the way, you are at the mercy of the other boards.

With Notches, we're trying to build a boat - something unique that has built-in barriers to entry, is not easily duplicated, and an ecosystem that generates that gravity. At the same time, we recognize that we’re not going to build the boat overnight and the only way we can gain the traction to get to the point where we have gravity is by having a presence on the sites that do have a lot of users and traction today.

-Tim

Join us at the nextNY Technology Primer for Business Folks

Do you have an idea but don't know where to get started on the technical side of things? Corey and I are helping organize a nextNY event intended for non-technical entrepreneurs, project managers, and so on. (Technical folks are free to join us as well, but we intend on keeping it very high-level). We will be joined by Charlie O'Donnell and Alex Lines, CEO and CTO respectively of Path101, and Ed Costello, a technology consultant.

This will be a very interactive and informal discussion to let you know what you need to get something built, how it should be built, things to consider when hiring a CTO or development talent, and just generally how to navigate the sea of buzzwords and frameworks and the like.

This event is on Monday, February 23rd at 6pm, and graciously hosted by Sun Microsystems (101 Park Ave at the corner of Park and 41st).

More details here – just note that if you are planning on attending you must RSVP for access to the building. Hope to see you on Monday.

-Tim

Alleged Extortion and Conflicts of Interest

A California newspaper made some pretty bold claims about unscrupulous business practices at Yelp, suggesting that maybe someone is taking this "PayPal Mafia" a little too far.

"Hi, this is Mike from Yelp," the voice would say. "You've had three hundred visitors to your site this month. You've had a really good response. But you have a few bad ones at the top. I could do something about those."

This wasn't your average sales pitch. At least, not the kind that John, an East Bay restaurateur, was used to. He was familiar with Yelp.com, the popular San Francisco-based web site in which any person can write a review about nearly any business. John's restaurant has more than one hundred reviews, and averages a healthy 3.5-star rating. But when John asked Mike what he could do about his bad reviews, he recalls the sales rep responding: "We can move them. Well, for $299 a month." John couldn't believe what the guy was offering. It seemed wrong.

Because they were often asked to advertise soon after receiving negative reviews, many of these business owners believe Yelp employees use such reviews as sales leads. Several, including John, even suspect Yelp employees of writing them. Indeed, Yelp does pay some employees to write reviews of businesses that are solicited for advertising. And in at least one documented instance, a business owner who refused to advertise subsequently received a negative review from a Yelp employee.

For its part, Yelp denies the accusations and questions the credibility of the anonymous sources used.

Regardless of the veracity of these claims, the story does highlight an important issue about credibility online reviews, credibility and conflicts of interest. The simple fact is that a business built on selling advertising to the subject of those reviews creates a inherent conflict of interest. If you make money selling to the same people you are reviewing, there is always going to be the temptation to certainly that temptation to reward "friends" or punish those who chose not to advertise. It's a pattern we've seen emerge whether it's editorial game reviews or user-generated reviews of restaurants. Though I personally don't believe Yelp would be so brazen, the problem is that you can certainly see how and why it would work, right?

This is one reason we've purposely avoided trying to build a destination site based on advertising, and it's also something we always think about when trying to craft incentive systems for partners and consumers. Even perceived impropriety can make us question the credibility of reviews – which ultimately dampen their utility.

Ultimately, we hope that our model and approach will enable us to stay more independent and trustworthy.

-Tim

A new look for the new year

As you may have noticed, we did a fairly major refresh on our visual design over the past weeks.

The old look was somewhat dated and, more importantly, cluttered. Our theme with the refresh was to simplify – aside from a new design, we also we spent a lot of time removing things. We’re still not done, but it’s a start.

You may have also noticed some logos for Notches Wire popping up around the site. Wire is a new product that we will be launching soon to making integrating reviews functionality, and it’s this product that made us rethink some of the visual design on the rest of the site. We’re currently testing Wire with a few partners and looking for a few others, so please sign up if you want to help us. I will post more details here once we’re closer to launching publicly.

I hope you enjoy the new look, and definitely welcome any feedback you have.

-Tim

Unstructured Gold

imageSomeone on the NextNY list (a fantastic resource for New York technology folks) posted about the latest feature from Dopplr where they allow users to discuss and recommend things travelers would be interested in.

This is the type of thing we strive to enable, embrace, extend with Notches. We believe the best review experiences are contextual. This "travelers" context is a great example and is not served well by the main review destinations like Zagat/Yelp/Citysearch for a variety of reasons.

The challenge is that this forum is unstructured information. Posters just start typing freeform text like “check out La Esquina’s sandwiches”. Automatically determining that La Esquina is a particular restaurant in SoHo and drawing that out is a real challenge. We’re in the early stages of making this work reliably, but it is so important to the long term vision. We will eventually be able to pick up the entities (restaurants, services, etc) in these posts, show Dopplr users what the rest of the web thinks generally (so you don't have to open up separate tabs for Zagat & Citysearch) and feed the information back into the big pool of data we are cultivating.

There is a whole world of fantastic, unstructured review information out there. We talk often about how the best information about hifi equipment or triathlon gear is on community specific forums. There is already some good information flowing on Dopplr's site that is getting trapped in freetext – just waiting to be unlocked, distributed, indexed, profiled & aggregated.

-Corey

Reviews and the Video Game industry

Video games are one of the verticals we are planning to attack first for a number of reasons. First of all, video games are a big investment for consumers from a time or money perspective. The typical video game for a modern console costs around $60 right now, and the time spent playing a particular video game is going to far outweigh the time spent watching any movie.

At the same time, there is a lot of shady stuff going on with reviews in the video game world – and we want to help eliminate this.

In September, it came out that Richard Cain owned both a game PR company, TriplePoint, and a game review website, GameCyte. The GameCyte team was staffed entirely with TriplePoint PR employees, and GameCyte gave glowing reviews of games from Telltale Games which Richard was an investor in.

From the interview:

Don’t you think… um. Obviously, it looks like there’s a conflict of interest here, when you have former and current TriplePoint employees – who still work for you, the founder of TriplePoint – reviewing and evaluating games from companies TriplePoint represents. Do you have a response to that?

Um, I don’t… people used to be… when Jesse used to work for us – and when I say “us”, I mean TriplePoint – you’ll see he does not review any game that he worked on, any client that he had any affiliation with. Um, otherwise I think the reviews credibly stand on themselves. Some have been positive, Telltale’s were, most actually have been negative. His review of Myst DS, for one. They did a review of, um, I think it was Double D Dodgeball, it was very negative, so I mean, the editorial independence, I think, speaks for itself.

I’d say this is an obvious conflict of interest – and not only was the relationship not disclosed but, as Gary discussed in the article, there were apparent steps taken to try to cover up the relationship. It certainly makes it difficult to trust the reviews on a site like GameCyte.

More recently, there was news that Eidos was trying to delay any reviews of Tomb Raider Underworld that were less than 8.0 (out of 10) to artificially inflate their aggregate scores.

When asked why, the spokesperson said: “Just that we’re trying to get the Metacritic rating to be high, and the brand manager in the US that’s handling all of Tomb Raider has asked that we just manage the scores before the game is out, really, just to ensure that we don’t put people off buying the game, basically.”

Eidos was caught in a similar sticky situation about a year ago over a negative review of Kane & Lynch.

GameSpot editorial director Jeff Gerstmann ran a pretty scathing video and print review of the game as well, giving the title a low rating of 6. Then he was fired. 

"After Gerstmann's savage flogging of Kane & Lynch, a game whose marketing investment on Gamespot alone reached into the hundreds of thousands, Eidos (we are told) pulled hundreds of thousands of dollars worth of future advertising from the site," Penny Arcade reported.

Gaming is big business, and the game studios know that reviews are important and are taking steps to try to manage them. This is made easier since Metacritic uses a straight average and doesn’t account for the review patterns of these editorial sites. The studios are very aggressive about advertising and have shown they are not shy about using this power. When these advertising dollars are the primary source of revenue for these editorial sites, once can certainly see how this can manifest itself in ways that are not exactly the best for the consumer.

One of our key goals with Notches is to build review summaries and aggregate scores you can trust. While we’re working to include editorial reviews as well, we feel strongly that user-generated reviews are an important element of this. While gamers may not appreciate nuances in the review process like professional critics, they also don’t have these obvious conflicts of interest.

-Tim

The birth of Notches

Corey recently dug up the initial e-mail he sent me that led to the creation of Notches.

From: Corey Henderson
Sent: Tue 9/20/2005 2:44 PM
To: Tim Marman

The concept is to take the idea of ratings way, way out there in terms of what has been done before.

Think del.icio.us or upcoming.org or Google maps or wikipedia, but for ratings. Pretty much anything can be rated: products (obviously), articles, places, restaurants, music, movies, on and on and on.

Right now, if you want to get a restaurant recommendation you could go to menupages.com, citysearch.com, zagat.com, and God knows where else. Imagine if there was an internet-wide consensus for restaurant reviews that would aggregate these sources. The sites themselves would then use an API to display the ratings on their site with some proprietary touches. It really makes no sense for someone like menupages.com to try and build a recommendation database by themselves, yet they try anyway.

I think the time has come to democratize ratings of all types.

Interestingly, I had a very similar conversation with Alex just a week earlier over sake at Decibel. I thought it was interesting to post here because our original vision hasn’t changed much from that time, and it’s something we’ve thought a lot about in the past 3+ years.

-Tim

dot-org

image As we mention in our FAQ, the Url “http://notch.es” has nothing to do with Spain. It is a “domain hack”.

There were lots of good reasons for us to be comfortable with this choice: Notches is a service, not a destination site, so we don’t need consumers to remember how to type it in. Traffic comes via direct links and Google, which makes owning the .COM superfluous.

However, it is hard to communicate verbally and it confuses folks who are not used to domain hacks. Even the granddaddy of the genre del.icio.us has moved to just delicious.com.

So, henceforth Notches’ home will be http://notches.org

Besides being easier to work deal with, the domain suits us just fine and reflects our values.

P.s. the old domain will still work as-is for a few more weeks, at which point we’ll put in 301 permanent redirects. I.e. you don’t have to update your links, but if you see something funky this could be the reason.

-Corey

Headless!

Antartica 14Albert Wenger at USV wondered yesterday whether “a headless web service can be successful. By ‘headless’ I mean a web service that is all back-end and does not have its own front-end.” He goes on to talk through some of the repercussions.

Notches is a prime example of this. In 2 ½ years of development, countless hours of system design, nearly 100,000* lines of code and only about 10% of all that effort actually renders HTML. (See iceberg pic for visual metaphor.)

You can see where the problem arises: a REST API is a hard thing to demo. “See here, we get back perfectly formed XML + JSON, isn’t it beautiful?” – Just doesn’t work.

So in our quest to let a million review sites bloom (we can’t, won’t, shouldn’t build them all) we built our own front-ends as a way to demonstrate what can be built, how to build them, drive API requirements and seed the database. We have plans to make the code and lessons learned available to all, stay tuned for news on this. Embrace & extend what we’ve done, please!

And a perfect example of this dynamic is the case of our Twitter app and subsequent partnership with Microrevie.ws. See our previous blog post for the whole story.

* 266k lines total, 18k are blank, 68k in comments, 35k are code-gen, and 50k is in third-party libraries.

(image CC licensed via Flickr/guilleavalos)

Announcing the Notches Toolbar

From the start, we’ve always felt that reviews need to be delivered contextually and thus have emphasized partnerships over simply aggregating reviews. At the same time, we recognize that partnerships are not always possible for a variety of reasons.

To that end, I’m very excited to be able to finally announce the Notches Toolbar, a Firefox 3 extension that enhances your browsing experience by adding reviews to sites that don't have them. Of course, this functionality builds on the Notches platform, so any reviews written on other partners (such as Facebook and Microreviews) will also be visible through the Toolbar (and vice versa).

toolbar

BeerMenus without the Notches Toolbar:

beers_noreviews

BeerMenus with the Notches Toolbar:

beers_reviews

In addition to overlaying reviews, the Toolbar also allows you to write reviews while browsing the site.

writingreviews

As always, your feedback is very important to us. Our goal for releasing the Toolbar is to get your thoughts and opinions to help shape future versions and what sites we support. Share any ideas, questions, or bugs with us on GetSatisfaction.

As of today, we support BeerMenus and ProductWiki. We’re also planning on rolling out support for a number of additional sites including eBay and Craigslist soon (which we can do dynamically without the need to roll out new versions of the Toolbar). We’re also working on designer tools to allow the community to add support for new sites. If there are sites that you feel should be included, simply click the “I want reviews on this site” button.

Find out more or install the Notches Toolbar.

-Tim

Meet Bharat

I’m very happy to introduce the newest member of our team, Bharat. Bharat is a talented developer and has experience working with large-scale systems, having previously worked at Xanga and Microsoft. He hit the ground running and is currently working on an as-yet-announced product that will help us extend the reach of the Notches platform.

We feel very fortunate to be able to add such a gifted software engineer and all-around cool guy to the team.

- Tim

We’re hiring

We’re looking to bring on developers to be part of the core engineering team and that can contribute to the product in a myriad of ways beyond coding. We want people who can ask the tough questions and challenge us. We want people who are not afraid to take ownership over an area and really drive it forward.

Our core platform is written in C# / .NET, so familiarity there will help – but ultimately we’re looking for smart, ambitious people with a good background in computer science, algorithms, and so on.

Our offices are currently in downtown New York City (SoHo). We’re certainly flexible in terms of hours but we do want to spend as much time as possible collaborating in person – in other words, we’re not looking for offshore firms or out-of-town developers right now.

You can find a more detailed job description here. If you’re interested, please contact us.

Update: We've filled the open position for now, but will be looking to expand the team again shortly.

- Tim

Simultaneous Discovery and its impact on stealth mode

We’ve talked a lot about the anti-stealth movement here and on the nextNY list, and the topic has resurfaced again recently thanks to Brad Burnham’s post about the advantages of being open.

I noticed that, at least anecdotally, there was a correlation between how open entrepreneurs were with us and their ultimate success. Simply put the entrepreneurs who are aggressively open in describing their plans seem to do better than the ones who are cagey. There is absolutely no data underneath this observation. It is just my sense after meeting hundreds of entrepreneurs over 15 years as a VC. If it is true, it could be for lots of reasons. The more experienced an entrepreneur, the more likely they are to understand that ideas are rarely unique, but the ability to assemble a team and execute against that idea is rare. Perhaps they are just more confident, and it is confidence that is correlated with success. But recently, I have started to think that there might be something more going on.

Brad concludes,

[In] the markets we invest in, there seems to be a real advantage to being open. The best entrepreneurs in those markets cultivate huge networks of knowledgeable people and engage them actively to refine their ideas. The companies they build seem to share this characteristic, opening themselves by publishing source code and APIs: betting that they can thrive in an open ecosystem by being able to absorb, process, and capitalize on relevant information better than their competitors.

It has also become a common refrain that ideas are nothing without execution. As Evan put it on the AngelSoft blog,

Anyone can have an idea, but it takes a lot of effort, commitment, ability to execute, and brute force optimism to launch a company and succeed. People in a position to compete with you are few and far between, so if you're the right entrepreneur for investors to bet on, then have confidence and don't waste time worrying about who steals your idea. Its very likely that many other people already had the idea, but just didn't have the skill to act on it.

There may be more to it than just that, however. Malcolm Gladwell had an interesting article in the New Yorker that suggested that not only are ideas not unique, but showed that they are often “discovered” simultaneously by multiple parties (emphasis my own).

This phenomenon of simultaneous discovery—what science historians call “multiples”—turns out to be extremely common. One of the first comprehensive lists of multiples was put together by William Ogburn and Dorothy Thomas, in 1922, and they found a hundred and forty-eight major scientific discoveries that fit the multiple pattern. Newton and Leibniz both discovered calculus. Charles Darwin and Alfred Russel Wallace both discovered evolution. Three mathematicians “invented” decimal fractions. Oxygen was discovered by Joseph Priestley, in Wiltshire, in 1774, and by Carl Wilhelm Scheele, in Uppsala, a year earlier. Color photography was invented at the same time by Charles Cros and by Louis Ducos du Hauron, in France. Logarithms were invented by John Napier and Henry Briggs in Britain, and by Joost Bürgi in Switzerland.

For Ogburn and Thomas, the sheer number of multiples could mean only one thing: scientific discoveries must, in some sense, be inevitable. They must be in the air, products of the intellectual climate of a specific time and place.

Curiously, Corey and I had this exact situation about Notches. Only days I met him to discuss his idea, I was sitting around with Alex and we had the exact same idea. On the nextNY list, David Rose also gave more evidence of this phenomenon:

Indeed, at New York Angels we've seen that things tend to come in waves: one month we'll get THREE plans for premium rums, then then next month will be two real estate sales web sites and four social networks for creative professionals, and so forth.

Think about what this means this for a second: Not only are your ideas not unique and special, but it is likely that others had the same idea. Even if you hoard the idea, your competitive advantage is tenuous at best. Brad theorized that successful entrepreneurs are open (or open entrepreneurs are successful) because “every time they describe their ideas, they learn more than they reveal, no matter how much they reveal. And, as a result, they are able to concentrate insight in a way that creates a defensible advantage for them.”

 

Perhaps there is another element at play. When determining whether an invention satisfies the obviousness requirement of patent law, one of the key tests is whether it has been done before. One reason for this is that, by and large, nearly every invention seems obvious once revealed – so the argument is that if it were truly obvious, someone would have done it before.

So, if you operate under the assumption that the idea itself is not unique, then making it freely available may in fact discourage others from executing on it because it creates the impression that the idea itself is in fact nothing special.

Thinking back on when I laid out the reasons that we were not anti-stealth with Notches, I think this was a big part of it. We never though of the idea as the key asset, but we didn’t want to make the idea freely available until we were in a position to be able to execute on it. We weren’t quite ready in November when I wrote that, but we are now – and thus we’ve been a lot more open about the vision.

- Tim