Orchive Blog

The road leading up to launch was an experience in itself. Productivity introduced itself to us in leaps and bounds, appearing to drift away and return at random intervals. Yet, in as the weeks closed down on our launch date, our team came together as one – the amount of work we got done together was astronomical. I am proud to have been a part of this team in its setbacks and success, our hard efforts culminating in an extremely successful launch.

The challenges we faced were not of a simple nature – they adapted, growing from an occasional itch to a constant swarm of bees buzzing around our heads, constantly alerting us of another issue that needed to be fixed. Development, the sector of our team that handled the physical creation of the product, encountered the most issues (naturally) as they dealt with various back-end and front-end difficulties. A day did not past where Dev did not encounter a new problem. Since the beta, we revolutionized the design and features of the product, forcing Design to come up with new creations for every aspect of the site. Marketing also faced a heavy task – we had to find a way to reach a large user base, and we used every resource possible. We analyzed our options and embarked on several campaigns, using social media, local news agencies, and interpersonal connections to maximize our influence.

The day that we launched, we utilized a social media to alert as many people as possible. Through, the combined efforts of the members of the team, we managed to get 150,000 hits on our page within the first 24 hours, exceeding our expectations. The feedback that we got from our users was invaluable – we’re currently in the processing of incorporating their advice and we plan to update the site in the near future.

We would like to thank all of our supporters who have been vital to the success of this company. Spread the word, and keep reporting!

Andrew Ng

Director of Product

“Everywhere, we see that power is shifting from large institutions to motivated individuals. Technology allows people to leverage the weight of these institutions against themselves, ” - Fareed Zakaria

Zakaria’s statement stuck with me when I first started considering the potential of Orchive. I’ve told countless individuals the same story about my observations during the Arab Spring and its relation to what Zakaria seemed to be hinting.

For those of you who followed the news during that tumultuous time period, you may or may not have seen the vast amount of pictures that media networks such as CNN, BBC, FoxNews and others were streaming to your television and computer screens. The significant deviation for these companies from previous international events was that, due to a lack of capability to enter some of these countries, they relied on different sources to obtain visual and written aid to understand what was occurring on the ground level. These sources were people, and the visual aid came in the form of pictures uploaded to Facebook and videos uploaded to Youtube. For the first time, instead of seeing photos taken by reporters, we were seeing photos taken by independent individuals. Even written reports were being given by individuals in the region.

Although the governments of these countries attempted to cut all communication–including an attempted “internet blackout”–word, picture, and video got out to the rest of the world. While critics later described the term “Facebook Revolution” as being overused and insensitive to the individuals involved in the actual revolution, there is no denying that the use of Facebook and other social networks to spread information, live from where it occurred, significantly changed the way the media was covering these events.

Since the rise of the internet, our world has become immensely globalized and interconnected. A man living in San Francisco can find out about an event that occurred in Xi’an minutes after it happened. A woman living in Mexico can hear about a fire in Germany before the German government can even release a statement regarding the fire. The point is, yes we are more globalized, and yes we are more interconnected than ever before. But the final evolution which we should now consider is the one Zakaria emphasized in his statement.

Power is shifting, from institutions –such as large media companies–to individuals. Whether Zakaria (a writer for both CNN and the New York Times) meant to imply what he did in his statement or not, it seems that the world is evolving so that the media companies become less relevant. As individuals begin reporting news themselves via social networks, news media companies will begin to take a middle-man position of simply connecting the reader with the reporter. At a certain point, you could ask yourself, why is the middle man even needed?

 Technology empowers us to do things we never imagined doing in the past. The internet has allowed us to take charge in global events mere moments after they occur. As media companies are criticized for being bias, subjective, or adding their spin on events, users will look for a different way to receive information.

Be part of the news evolution, become part of Orchive. 

Posted by Ofry Shatzky

Orchive’s approach to comments is going to be something like this: while traditional news websites base their comment ranking by date, Orchive is going to allow every comment to get a merit-based opportunity to shine. it’s not just early-bird-gets-worm. 

Let’s say there are 3 comments. Comment 1 has a score of 2. Comment 2 has a score of 2. Comment 3 has a score of 1.

Comment 1 and comment 2 will ALWAYS be above comment 3. However, comment 1 and comment 2 might randomly switch places. This is to give an equal opportunity to both comments.

I originally thought to use a naive method to randomize it. Simple: just search for comments that have matching scores, and then randomly switch places. This method is pretty decent, but it scales horribly. The more comments you have, the worse it gets. The next method, on the other hand, is efficient and blind. The program doesn’t need to know the scores in order to execute these instructions! How awesome is that? The algorithm is as follows:

1. score = upvotes minus downvotes
2. score = score + (random_number_from_zero_to_one - 0.5)

And that’s it. It’s so simple but it works *exactly* the same as an algorithm that isn’t blind — that keeps track of the score. 

Posted by Jonathan Tang

Standard news websites do well by sorting their articles chronologically, with perhaps a bit of human intervention to push particularly relevant articles to the top. Though websites like FMyLife have tried, this manual-sorting system doesn’t work well with websites based off of user-submitted content like Orchive.

Trendy submissions on YouTube get almost no help at all from the website itself. Videos can become viral only through manual searches, recommendations from the YouTube sidebar, and perhaps front page status — if they’re lucky. Videos with many views but terrible ratings are listed nevertheless, as if they’re hot, in the YouTube sidebar. Perhaps the joy in YouTube is searching for videos that pertain especially to the user’s interest; however, the shallowness of the YouTube frontpage attests to how uninteresting such a system can become. However, there are advantages to such a system — unlike Orchive, YouTube is intended for personal, individual viewing.

Urban Dictionary’s system judges submissions only on a score based off of thumbs up and thumbs down. The system seems simple and intuitive, but any of us who’ve actually been on Urban Dictionary can describe how immutable the “definitions” are. Making submissions on Urban Dictionary is only a matter of being the first one in. Once a definition is established, it’s nearly impossible to knock it down as users will inevitably vote for the one at the top.

The hotness algorithm on Reddit accounts for user-submitted voting and time, while also using a logarithmic scale. Such a system means that newer submissions generally have precedence over older ones. While submissions do go viral, they never get too many points as to linger longer than a few days — a trait important on a website that relies on aggregating interesting web links. 

While I did draw influence from Reddit in designing our algorithm, our team decided that pageviews should also play a minor factor in determining the “hotness” of a post. 

Orchive’s trending value algorithm is designed to present good, bias-free journalism to the user on a silver platter. This algorithm, combined with a sort of natural selection from user-submitted votes, ensures fresh, top quality posts making it to the front page. Our goal is to allow the users and website to sort themselves naturally, with administrators never to intervene in promoting submissions.

Posted by Jonathan Tang

Allowing users to sort through a messy sludge of posts is one of the problems that we face when developing a website like Orchive.

eBay and Amazon’s reputation-based system works to filter out scammers. Given Orchive’s mission — to completely eradicate the editorial bias of modern journalism — we needed to take that into consideration when developing Orchive’s Newsworthiness system.

Similarly, Reddit’s “karma” counter encourages users to submit many posts, which are filtered through user voting. However, we sought to keep Orchive’s system to a decent limit, in order to prevent power posters from developing. Reddit’s political discussion forum is notoriously plagued by sensationalized headlines, a problem we’ll address by allowing posts to be flagged.

Newgrounds is different from Reddit and eBay — instead of making user points based off of submissions, the developers on Newgrounds allow users to deposit “experience points” daily, in order to gain “levels”. What ends up happening is that users continue going back to the website to deposit their experience points. Users proudly display their levels on their profile pages and the Newgrounds forums. Orchive, however, will try to keep posts free from bias by only listing such achievements on userpages.

We specifically avoided drawing influence from websites like Twitter and 4Chan. Although Twitter is fine for small-time social networking and following celebrities, using Twitter’s search system to find relevant posts is bothersome at best. 4Chan is even worse — the lack of a sophisticated username system, combined with a messy forum layout, makes posting difficult, messy, and unrewarding (though the idea of almost complete anonymity draws many users back to the website).

I ended up devising a system like this: a given submission is assigned a value of reputability that is based off of pageviews, positive votes, negative votes, accuracy, and total voters. This value is added to the original submitter’s reputability count, which also determines the user’s newsworthiness. I haven’t sorted out the details yet, but good comments should also be linked to a higher newsworthiness. Algorithmically generated milestones will promote the user after he or she reaches a certain reputability value.

Along the way, the user can also achieve trophies (which are difficult to attain) and medals/ribbons (which are easier to attain) based off of reputability and other criteria. 

The main philosophy that goes into this code is this: users should be individualized and rewarded for good submissions. There should never be a concern that having too many people will dilute a user’s voice. 

Following along with this, an idea I’ve been considering is having article comments being listed randomly. The comments will still be rated, but comments with similar ratings will be randomly grouped with each other. This ensures that every comment has a chance to be heard, not just the earliest ones.

Posted by Jonathan Tang

Perfect form;

The ancient Greek philosopher Plato discussed perfect forms. Ideally, he explained, perfect forms are unreachable and unobtainable by normal humans. When asked what “things” could have perfect forms, Plato responded that everything could have such a form. When asked to list some things, Plato stated that objects, emotions, concepts, feelings, ideas, and even words could have perfect forms. Plato went on to say that as humans, we will never reach nor fully comprehend what a perfect form of anything is.

Looking back, I always find Plato’s perfect form theory rather interesting. Whether or not one agrees with Plato, I think his theory deserves to be considered through a contemporary lens. Every time an event takes place in human history, as soon as it is recorded, it begins to trail off from being recorded as it actually occurred.

While no one can trace who originally said this quote, contemporary U.S. historian Howard Zinn is often recognized to having said “History is written by the victors”. In his eyes, and in the eyes of many historians, history will always have a bias because history is always written from a certain perspective (that of the victor). Every event that unfolds will be viewed in a subjective manner. As such, the actual event, objectively, may never be fully understood. Plato’s words begin to make sense. The perfect form of history will never be learned. The perfect form of day-to-day events will never be heard or recorded because everyone records a story with a certain bias.

Roughly around the time when Orchive first started, I presented to Francis, co-founder and CEO of Orchive, what I called “Plato’s Corollary”. Plato’s corollary was what I concluded, the next logical conclusion one may be able to make from Plato’s perfect form theory. Although, as humans, we might never be able to visualize the ideal perfect form of anything, if we obtain enough different perspectives on one single “thing” we might eventually come close to understanding the true perfect form. For example, one who has seen only one type of chair in his life time we always visualize that chair whenever he considers the word “chair”. On the other hand, a person who has seen chairs from all different countries and all different cultures might obtain a very different image in his mind when he considers the same word.

Along comes Orchive. We don’t promise that we will have one single report which will accurately tell you what is going on anywhere around the world. We can promise you that in our vision, by aggregating a collection of reports pertaining to one single issue, you, as the reader back home, may be capable of forming the best understanding of an event.

Plato might have believed we may never understand perfect forms, and he might well be right, but through Orchive, through a multitude of different reports from different individuals, we hope to obtain the best objective understanding of the events that take place worldwide.

Posted by Ofry Shatzky