About 6 years ago now, in 2011 I had this cool idea; rather than just the celebrity tweets and the occasional news articles which were being posted to Twitter at the time, wouldn't it be great to post news stories and then discuss these stories online via twitter with other users?
The following is a small post mortem on VocZie, the idea, it's impact and the work that would be needed to bring it up to date.
The Hope : Sharing views on news will allow me to learn more
Imagine reading a new story on your favourite news site and at the same time seeing a live panel of Twitter comments from others who were discussing the article.
In terms of social change, I had this idyllic, and a little naïve view that by sharing views it would allow us all to better understand the stories we were seeing in the news. I recorded the following thought in my notes at the time:
So much of the news is curated and presented to us by news organizations and algorithms that we often don't get the whole story. There are still times today when I'll ask a friend about a news article and get a completely different view on the same story. If this experience happens to me in real life, then perhaps by extending how I read news I could get a wider view online. I could share my views on a news article and have a conversation with the whole world."
The Reality in 2011
In 2011 when I launched the site Twitter wasn't the source of news sharing it has become today. At the time it had a limit of 120 characters per tweet, and that included the URL of the news story. The result was that most of the news links that were shared where shared without comment, simply a link to a story. The discussion of viewpoints and ideas I had been hoping to discover just didn't exist.
Then, just as I was completing the site, news came from twitter that they were limiting the Twitter API. In doing so they wanted to stop people from using none-twitter owned user interfaces. These changes would basically cripple my site and prevent it from having the number of users that I would ultimately like to see. It was the death nail in my idea.
Playing in someone else's garden
VocZie had a hard dependency on Twitter, and Twitter's ecosystem. When you are building within a closed ecosystem like Twitter, you are ultimately at Twitter's whim. If Twitter changes its terms of service you can be out of business overnight. There are opportunities to make money, but you need to be careful when and how you invest in them. I was pretty devastated when the API change was announced by Twitter. But realistically I had even bigger problems.
Dude, where's my business model?
VocZie was a fun idea, but it had no business model. I took it on as a learning exercise and as a chance to explore social and news interaction on line. It never had a business model, or an idea about how it would generate any cash. Sure, the site might generate value for people, but would they pay for that value? Would enough people be interested in the value that VocZie brought to news reading that it would allow me to generate cash via advertising? - All great questions, but ones I ignored at the time. However, keeping the site going is a costly experience. Not only are there hosting costs, but there are also time and maintenance costs. The site needs updating, everything from the version of PHP it is running on, through to the VM and the virtualization technologies that my provider supplies; they all need updating.
The Reality in 2018
News and Twitter have both changed dramatically since 2011. The rise of "Fake News", and indeed the rise of use of Twitter and Facebook for political discussion and advertising has really changed the scene. People have become more polarized than before. Arguably there is a need for something like VocZie.com to help bring people closer together and to encourage debate and ultimately better understanding of each other. However, for this to happen VocZie.com would need some significant changes...
Bringing VocZie up to date
There are some technical limitations on how VocZie was implemented, but that is a product of my learning experience. To make VocZie work for the new reality of news and twitter would require the following changes.
VocZie currently uses RSS feeds (news feeds) from news websites. When you click a story, it will highlight the associated tweets. However, news providers and audiences are becoming ever more polarised. Fox news fans aren't fans of CNN, and Guardian fans don't read the Daily Mail. But for discussion to occur then equivalent reporting between the news sites needs to be matched, and then the comments aligned. So that as a VocZie user you can understand the full gamut of views both by the press and by their readers. Implementing this will require some natural language processing to ensure that stories about the same issues are matched across news sites. Then a large rise in the search terms to try to catch all tweets about all of the related news feeds.
Twitter now has the concept of a conversation with users able to reply to tweets. This wasn't included in the original implementation of VocZie.com. Instead VocZie.com simply found the tweets which mentioned the URL of the news story. Additional work would be needed to identify and show any replies to those tweets and the associated conversations. Indeed, we'd want to reply to these conversations and that would need to be supported by VocZie.com as well.
A common approach used today when folks have a lot of information to share on twitter is to create a "thread" or a series of tweets in which the original tweet author effectively replies to him or herself. This concept is supported by Twitter, but not by VocZie.com and additional work would be required to add this, both for reading the content posted to twitter and also for allowing extra-long tweets.
Same subject, no URL
As I discovered back in 2011 people will often post about a news story or political item on twitter, but this usually doesn't include the news story or a news source. These comments are often more insightful than the original tweets which mention the story URL, however they are missed by VocZie.com. Similarly, hashtags associated with a story are also excluded and could provide additional context for a conversation.
Supporting Twitter's longer tweet length (256 characters) is important too, and VocZie.com would need to be updated to support this.
All of the above will require a large UI update and change, and that excludes some of the technical challenges which are yet to be addressed.
One of the unique features of VocZie.com is that not all of the work is performed by the server. Searching twitter is actually performed by the user’s web browser and the results are then transmitted back to the server for other users. This was done to get around twitter's search cap, Twitter limits the number of times a given IP address (computer) can call it's search service. VocZie.com is a search heavy site. To address this the search functionality was moved to the client. This works, but it can make the site seem slower than it should.
PHP and Migration
The VocZie.com code base needs to be moved from its current PHP implementation to a new version of PHP. Even after this is complete the hosting provider I am using is shutting down the data centre that hosts VocZie. I’d need to migrate the entire application to a new type of virtualisation technology in a new data centre.
Finding the Money
With all of these changes, VocZie.com will need to make money, to pay for the service and its development costs. So much of the news content and social media is free to the end user. I'm unsure if anyone would be willing to pay for a site which presents them with news reported in a style they may not like, along with opinions which they may disagree. It all seems like Monty Python's Flying Circus' sketch "Argument Clinic".
I'd love to think people would pay for such an experience, but I'd doubt it. So, in the end I think the time has come to pause VocZie.com. I'd love to come back to it one day, and I might yet. But for now, I think the time has come to hit the big red the power switch in the cloud, and turn VocZie.com off for now.