It's intended for beta users, potential investors and other interested parties. It should only take 5 minutes to go through, and should help to describe what we're about for all those who are still confused or unsure of Know Between's potential. Enjoy.
Thursday, 12 June 2008
Overview presentation posted
We released a presentation today that gives an overview of Know Between, what the vision is, what you can do currently with the beta, and what we need to take it further.
It's intended for beta users, potential investors and other interested parties. It should only take 5 minutes to go through, and should help to describe what we're about for all those who are still confused or unsure of Know Between's potential. Enjoy.
It's intended for beta users, potential investors and other interested parties. It should only take 5 minutes to go through, and should help to describe what we're about for all those who are still confused or unsure of Know Between's potential. Enjoy.
Tuesday, 27 May 2008
A customer proposition for Know Between
This proposition is both a pitch and vision for the customer value of Know Between. It's broad enough to endure all the directions in which Know Between may head, yet compelling enough to get people thinking about specific applications, and the explosive potential of Know Between.
The customer proposition is:
Demonstrate your credibility online by answering questions, posting whitepapers, holding discussions, receiving recommendations, achieving awards, knowing people and more.
People you can trust are your friends, friends-of-friends and experts.
Valuable people are those with a high ranking in one or more topics you’re interested in.
Find valuable people you can trust by connecting with friends-of-friends and experts, who are incentivised to strengthen their credibility online.
The customer proposition is:
“Demonstrate your credibility online and find valuable people you can trust”Your credibility is a list of all the topics you’re interested in, with your expertise ranking relative to others.
Demonstrate your credibility online by answering questions, posting whitepapers, holding discussions, receiving recommendations, achieving awards, knowing people and more.
People you can trust are your friends, friends-of-friends and experts.
Valuable people are those with a high ranking in one or more topics you’re interested in.
Find valuable people you can trust by connecting with friends-of-friends and experts, who are incentivised to strengthen their credibility online.
Distributing Know Between to other websites
Know Between already has a number of 'hooks' outside of the website itself. You can get email notifications, RSS feeds and Twitter updates. You can even view and share your expertise on Facebook, with the Know Between Facebook app.
Using these hooks, other websites can make use of Know Between's powerful features. For example, Openmind projects - a leading volunteer organisation in Thailand - shows recent news (questions, responses or people) for the topic Development.
The Openmind example is just read-only of data and not personalised in anyway. Imagine accessing and developing your Know Between network through eBay for example. If a friend or friend-of-friend has sold lots of electronics and got good feedback for it, then they might be a good person to trust for buying a second-hand digital camera.
Know Between's true power will be unleashed when the features it offers are available in digital the places people use everyday.
Using these hooks, other websites can make use of Know Between's powerful features. For example, Openmind projects - a leading volunteer organisation in Thailand - shows recent news (questions, responses or people) for the topic Development.
The Openmind example is just read-only of data and not personalised in anyway. Imagine accessing and developing your Know Between network through eBay for example. If a friend or friend-of-friend has sold lots of electronics and got good feedback for it, then they might be a good person to trust for buying a second-hand digital camera.
Know Between's true power will be unleashed when the features it offers are available in digital the places people use everyday.
Importing your existing networks
One of the biggest barriers with new social or professional networks is that users typically have to start building their friend lists again. People object to doing this. They have become loyal to their email address book, Facebook friends list or LinkedIn connections.
Know Between of course supports 'connection growth' through question interactions, exposing friends-of-friends, and direct invite (if you know someone's email address). But people still have to build their digital networks up in yet another application.
All the big social networking sites try to get you to find more friends and build your networks. This of course helps to provide more relevant and useful content and services, and promotes loyalty. Asking for your email username and password is a common approach, but many people are concerned about the privacy implications of that.
In May, MySpace, Google and Facebook announced plans to export social data for use in other websites. LinkedIn is also opening up their data to specific websites. This effectively levels the field for any existing or new applications that want to use social networks, even though the amount of control or access to that data may be limited. Still, great news for Know Between.
Know Between of course supports 'connection growth' through question interactions, exposing friends-of-friends, and direct invite (if you know someone's email address). But people still have to build their digital networks up in yet another application.
All the big social networking sites try to get you to find more friends and build your networks. This of course helps to provide more relevant and useful content and services, and promotes loyalty. Asking for your email username and password is a common approach, but many people are concerned about the privacy implications of that.
In May, MySpace, Google and Facebook announced plans to export social data for use in other websites. LinkedIn is also opening up their data to specific websites. This effectively levels the field for any existing or new applications that want to use social networks, even though the amount of control or access to that data may be limited. Still, great news for Know Between.
Browsing people, topics and questions
It's hard to get a view on who all the people are on Know Between, what all the topics are and what all the questions are.
If you're not logged in, you get to see how many topics, questions, responses and people there are, but you can't click to browse them.
If you are logged in, you can see your friends and friends-of-friends (who share interests), but again there's no holistic browse.
The search feature indexes topics and questions but not people. It would be good if you have a friend called Eric to just search this rather than trying to find the link hidden somewhere.
I suggest the following updates:
If you're not logged in, you get to see how many topics, questions, responses and people there are, but you can't click to browse them.
If you are logged in, you can see your friends and friends-of-friends (who share interests), but again there's no holistic browse.
The search feature indexes topics and questions but not people. It would be good if you have a friend called Eric to just search this rather than trying to find the link hidden somewhere.
I suggest the following updates:
- Display the number of topics, questions, responses and people always (whether logged in or not)
- Support browsing of all topics, questions and people
- Show people in search as well as topics and questions.
Saturday, 24 May 2008
Tag hierarchy
There's been some discussion about the effectiveness of the topic hierarchy on Know Between.
The topics are designed to associate questions with relevant subject areas, so that they can target people who might be interested in or able to answer them, and award credibility accordingly.
Open tags
Initially, topics were just a sea of open tags. Users were able to choose whatever tags they wanted without further questioning. The drawback was that there was no relationship between tags and a lot of duplication. For instance, 'UK' and 'United Kingdom' were different; 'Birds' and 'Oystercatchers' were not connected in any way. The whole concept of building credibility and ranking for topics was in jeopardy.
Fixed hierarchy
A fixed hierarchy is commonly adopted to overcome the problems of open tags (see Yahoo! Answers and LinkedIn). The drawback here is that you're forced to limit yourself to a predetermined set of topics. There's no room for dynamic adaption of the hierarchy.
Infer relationships between tags
Another way to realise the benefits of a hierarchy or relationships without enforcing fixed categories is to provide some sort of smart, automatic matching. Flickr uses 'clusters' to establish related tags, and they provide an API to this feature which Know Between uses.
Auto-tagging
Yahoo! offer a service to extract significant words or phrases from a larger piece of content. The Term Extraction service may be useful at identifying probable tags from a question, although it will never return words that aren't in the piece of text.
Dynamic hierarchy
For Know Between we have chosen to put in place a system whereby users can choose their own tags or browse an existing hierarchy. If not browsed through the existing hierarchy, users are prompted to confirm the parent category for each tag that matches an existing one. For new tags, users are prompted to choose where in the existing hierarchy the new tag should be placed.
An initial top-level category list starts things off. These cannot be selected as topics, only as category parents. This has been a bit confusing for some, but the intention is to ensure more specific tags are chosen. Over time there will be a large set of topics that have dynamically been added. There will be fewer and fewer new topics, so less of a barrier for people asking questions.
How do you think the tagging of topics to questions can be improved?
The topics are designed to associate questions with relevant subject areas, so that they can target people who might be interested in or able to answer them, and award credibility accordingly.
Open tags
Initially, topics were just a sea of open tags. Users were able to choose whatever tags they wanted without further questioning. The drawback was that there was no relationship between tags and a lot of duplication. For instance, 'UK' and 'United Kingdom' were different; 'Birds' and 'Oystercatchers' were not connected in any way. The whole concept of building credibility and ranking for topics was in jeopardy.
Fixed hierarchy
A fixed hierarchy is commonly adopted to overcome the problems of open tags (see Yahoo! Answers and LinkedIn). The drawback here is that you're forced to limit yourself to a predetermined set of topics. There's no room for dynamic adaption of the hierarchy.
Infer relationships between tags
Another way to realise the benefits of a hierarchy or relationships without enforcing fixed categories is to provide some sort of smart, automatic matching. Flickr uses 'clusters' to establish related tags, and they provide an API to this feature which Know Between uses.
Auto-tagging
Yahoo! offer a service to extract significant words or phrases from a larger piece of content. The Term Extraction service may be useful at identifying probable tags from a question, although it will never return words that aren't in the piece of text.
Dynamic hierarchy
For Know Between we have chosen to put in place a system whereby users can choose their own tags or browse an existing hierarchy. If not browsed through the existing hierarchy, users are prompted to confirm the parent category for each tag that matches an existing one. For new tags, users are prompted to choose where in the existing hierarchy the new tag should be placed.
An initial top-level category list starts things off. These cannot be selected as topics, only as category parents. This has been a bit confusing for some, but the intention is to ensure more specific tags are chosen. Over time there will be a large set of topics that have dynamically been added. There will be fewer and fewer new topics, so less of a barrier for people asking questions.
How do you think the tagging of topics to questions can be improved?
Tuesday, 13 May 2008
What's this blog for?
This blog is for comment, discussion and news about developments of Know Between. For more about Know Between, see my overview post.
In this blog, I'll be introducing you to changes that are happening or planned to happen, and things you can do with Know Between. I'll inform you of any problems or limitations I'm aware of and start discussions about certain features or ideas. I'll be inviting guest bloggers to join me also, and of course you can (and should) comment on any posts initiated here.
I also have a sister blog - The Trusted Web - that covers more general discussion of the growing need for credible knowledge sources online, for which Know Between is helping to address.
If you have questions about using Know Between, or want to suggest possible enhancements, you may like to use Know Between itself. There are topics already set up for Enhancements and Using Know Between. Are you an expert in Know Between?
In this blog, I'll be introducing you to changes that are happening or planned to happen, and things you can do with Know Between. I'll inform you of any problems or limitations I'm aware of and start discussions about certain features or ideas. I'll be inviting guest bloggers to join me also, and of course you can (and should) comment on any posts initiated here.
I also have a sister blog - The Trusted Web - that covers more general discussion of the growing need for credible knowledge sources online, for which Know Between is helping to address.
If you have questions about using Know Between, or want to suggest possible enhancements, you may like to use Know Between itself. There are topics already set up for Enhancements and Using Know Between. Are you an expert in Know Between?
Subscribe to:
Posts (Atom)
