If there's one thing to be learned from social media tools, it's that these services were not made to interact with one another. Complaints are rolling in and heated discussions are taking place about the noise levels within social media platforms. Here's a look at why noise levels are high and why filtering should be the next step for social media platforms.
With so many different platforms to aggregate, noise levels are surging. An underlying issue in the level of noise is that some of these services were not made to interact with one another. Users of social aggregation tools should understand that what you may consider noise is actually a side-effect of using a social aggregation platform. Users should also note when you may be confusing aggregation with importation.
Though importation may be a necessary step within aggregation, there is a difference between the two. Importation is usually more selective and limited to the inclusion of select items of other services. This may include common specifics such as names, ages, and locations. However, with aggregation a service attempts to conglomerate key features and entire contents of other services. This makes aggregation seemingly more flexible due to it's ability to encompass a wider scope of content.
The services that are aggregated are usually used for entirely different purposes. For example, some platforms are used to keep in touch with others such as family, friends, or business contacts. On the other hand, you have services that are used only for the purpose of finding more content and conversations pertaining to certain contents.
When you pull in an account from a platform that is completely unrelated to to the usage of another, you will inevitably create a small amount of noise. However, with social aggregation platforms it's hardly ever just one account. This can increase the noise level to an irritating high for other members of these services, including those within your personal network.
Aggregated services are not only used for entirely different purposes, but also cater to different audiences. Consequently, who you may be catering to is dependent upon why you may be using the service. While some articles or content submitted to services may overlap, this is only because there are overlapping interests for the different audiences on these platforms.
How does this affect noise levels? If you're using a service to promote content, you may be considered noisy to those that are looking for conversations. If you're using a platform to keep in touch, then those looking for content and in depth conversation surrounding particular content would need a way to block out idle chatter.
Filters are rapidly becoming a pertinent issue for developers of social media services. As a result, social aggregation platforms are in the perfect position to lead the pack. While this is no easy task and one that cannot be solved in its entirety, it would help resolve another issue social media users are facing: courtesy.
Instead of being able to freely add whatever service you wish, some users like myself are taking into account what others may consider noise on certain services as a courtesy to members. In essence, you are becoming our own filter. You may refrain from important other services for fear of being labeled as "noisy". With better filtering options, users can use these services to their fullest extent without becoming a nuisance to others or missing the benefits of aggregating all of their accounts.
Listed below are links to blogs that reference this entry: Why Filtering is the Next Step for Social Media.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/3934
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
Great post Corvida. Couldn't agree more, as social aggregation continues and more services are added. There must be new ways to categorize, filter and manage the barrage of information. thanks for the insightful post.
Posted by: Steaprok | May 10, 2008 3:48 PMfiltering, importation, aggregation
check...
Posted by: neurokinetikz | May 10, 2008 3:53 PMhttp://neurokinetikz.org
This definitely goes hand in hand with the trend we have been seeing of new vertical social networks popping up all over the place. Do you see the future belonging to platforms like facebook and myspace? or to smaller targeted niche vertical platforms. Eventually we are going to reach a saturation point.
Posted by: Jacob Morgan | May 10, 2008 3:56 PMvery insightful post. despite all the hype, social networks and web 2.0 sites are pretty broken when it comes to working together. we definitely need more filtering.
Posted by: Pedro | May 10, 2008 5:46 PMTo generalise a bit more, I don't think the special stuff is in aggregation, it will be in dataportability consumers that can filter based on the rules that you want.
Instead of aggregation we should have filtering capabilities in each service, or use filtering middleware.
Aggregators like Friendfeed are a stop gap for true DP. Friendfeed needs to position itself as a destination in it's own right not just an aggregator, which I think it's doing quite well. If all my friends weren't so active on twitter i'd be using friendfeed more. :)
Posted by: David Novakovic | May 10, 2008 6:21 PMDear,
I'm a korean living in seoul.(name: Oh Minseok)
Korean special investigative team investigated samsung
corporation.
But they did not work right.
Suspicious to have been bought off.
Samsung corporation had many crimes.
And the team investigated samsung corporation.
It contained korean companies samsung,huyndai,sk CEOs' illegal issuing
stocks or bonds. ( previous CEOs or present CEOs )
The quantity were plenty.
(Three company CEOs did(and are doing) many crimes to me.
Many koreans are knowing it.
But many koreans are bought off by illegal issuing stocks or bonds.)
The team knew it.
Korean special investigative team must investigated this.
But they concealed it.
I ask for asking for this criminal investigation to prosecutors in any
country.
And help the shareholders and me.
P.S)
Posted by: oh minseok | May 10, 2008 7:19 PMThree companies are hacking me and trying to kill me.
And are suspicious to use my name and email illegally.
If you receive another message that I dictated above are not true,
it is not from me, but from three companies.
The things I dictated above are true.
Actually, it is not the next step. The current situation is a step back to what was already there.
Filtering has been and always will be a highly personal matter and thus has to be done first and foremost on the side of the user - and it starts with better recognition of double content and ends with learning of the behaviour of the users actions.
But I hope the current flooding of all content everywhere will Decrease the time until the filterin mechanisms kick in an devs start going back again and discover how much filtering algorythms are already available if only they would implement them.
Posted by: Nicole Simon | May 11, 2008 4:57 AMIt would be much more interesting to see social networks/media and other web 2.0 technologies that allow users to choose what they want.
Otherwise...
Posted by: Loy | May 11, 2008 6:33 AMI really like how a service like FriendFeed brings it all together, though I agree that it can get noisy sometimes.
Because I follow/listen/commune with someone, should I be forced to see everything?
I'm not so sure I need to see every Pownce or LastFM update of everyone I 'friend' - yes, it is part of their online persona - what they are communicating. Therefore, if I've chosen to follow someone, I follow every bit of them.
Still, wouldn't it be grand to to turn off a portion to follow only the portion we want to hear. (see the Alltop sections).
For me, the signal of these aggregated services is ahead of the noise. I get more knowledge in less time.
Posted by: Mike Sansone | May 11, 2008 7:20 AMI think there are some quite simple solutions to the problem of Web 2.0 content filtering too (see e.g. oddflower), but most people are not really interested. The trouble is, filtering is inherently less social than simple aggregation, because the people as personalities would come to matter less than the information getting shared. As you suggest though, this would not only be more productive, it would be a lot more sociable as well!
Posted by: gav | May 11, 2008 8:12 AMIts my fear of causing unnecessary spam to others that actively discourages me from using, for example, Seesmic much.
If I want to get a Tweet out to only the Sysadmins that follow me, I prefix the Tweet with "Hey admins:" , but its still spam to all my other followers, and its hardly efficient targeting.
I posted more about how Services like Seesmic, Twitter, Friendfeed need to include at the very least group features, and I if not them, then at least client apps like Twhirl.
http://jemimus.net/?p=509
Posted by: Robert | May 11, 2008 8:21 AMMy current pet peeve is that adding photos to my Flickr account adds them both to my Facebook mini-feed and my Friendfeed which then also adds them to my mini-feed. The obvious solution is for FriendFeed to add some granular filtering capability but I don't expect to see it soon.
Posted by: Michael PateThe whole premise of the article is on stating that "these services were not made to interact with one another". Maybe not a year back. But now everyone realizes that it has to happen, and it will. Will love to see you write on this subject 6 months from now!
Posted by: Technology Watch | May 11, 2008 9:28 AM-Des
TechWatch
Interactive UI and filters features are social-weapons we may provide ;o) .oO(seem I'm on better way)
Posted by: Olivier Duprez alias ze kat | May 11, 2008 3:57 PMI completely agree that better filters are urgently needed. We are working on one such filter at www.socialrank.com.
Posted by: Mike | May 11, 2008 4:34 PM"filters are the future", "we need better filtering", "Filtering ... is a highly personal matter".
I think the answer to aggregation and filtering is in XTM (XML Topic Maps), a standard designed to describe knowledge using ontologies. Ontologies are portable topic maps that can be used as a lens to look at any data. XTM ontologies are like lenses and may seem like a "filter" but indeed it is a "map" not unlike the idea of using different Google maps on top of Google Earth data (e.g., you can easily see the data with streets, without streets, with any sort of geo-data or without.
Social networking information in an XTM format is [one] model that might help Web 2.0 data become more useful to everyone because ontologies can be personal, global, or focused for any domain of expertise desired. Ontologies can be merged and superimposed on any data that is XTM compliant.
At MyST we experimented with a very fringe (but practical) idea that was inspired by XTM and topic maps. We built a crude "topic cloud" ontology and deployed it on all of our client sites. The objective was to improve on the idea of a tag cloud - take a look at our topic cloud and also Sally Falkow's (she's an expert on social media). Make sure you drill into these to see the unique visibility and discoverability of the content.
bf
Posted by: Bill French | May 11, 2008 7:39 PMAnd yet possibly the most effective and advanced RSS filtering tool to date - BlogFriends, which launched 1 year ago - ran out of money in March this year. Gotta love that visionary UK investment scene
Posted by: JofArnold | May 12, 2008 1:32 AMCurses; Moveable Type filtered out the end of that comment! It was something like "stir stir stir" followed by a smiley. Probably filtered it out cos I used the greater than sign which I guess it thought was an html/script tag. Hopefully you get the point that I'm stirring and not at all bitter :-D
Posted by: JofArnold | May 12, 2008 1:35 AMCorvida,
Hope all is well. I do not completely agree. For me it's not a "filter" but more the "bait" I set out. This could be the same coin, but looked at from 2 sides ...
What I'm aiming at is the concept of of VRM where the user is in control. He will provide an input to a personal filtering system, by telling "him" what is interest is. From there certain filters kick in...
What do you think?
Bart
Posted by: Bart Stevens | May 12, 2008 5:43 AMBart:
"He will provide an input to a personal filtering system, by telling "him" what is interest is."
In XTM vernacular, you are describing a personal ontology that's superimposed on any information set. From that, emerges a "map" that embraces the personal interests showing what's "nearby" (or closely relevant and of interest) and what's far away (not that interesting or relevant).
Unfortunately, this is mostly theoritical (at a social computing level) as few social information resources are available as XTM. ;-) However, there are some good companies out there (such as Ontopia) providing tools that are well beyond theoretical stages of implementation.
bf
Posted by: Bill French | May 12, 2008 7:39 AMFriendFeed needs this like yesterday.
Posted by: Sam Jones | May 12, 2008 12:24 PMThere's an argument for filtering within the services themselves.
A pet peeve of mine is getting stuff sent to me on Digg and StumbleUpon that I don't care about.
I want to block all suggestions by friends that don't match the topics I subscribe to.
As for Yahoo! Pipes, I wrote about it over a year ago and I'm still surprised that it's not more mainstream.
Yes, it's a monster to figure out, but any number of mashups could have tamed that by now...
Posted by: Wayne Smallman | May 13, 2008 5:13 AMnice post.. i agree with you..
Posted by: Faisal | May 14, 2008 3:27 AMI've always thought that Bayesian filtering is a great way of tackling this stuff.
Posted by: gsiener.myopenid.com