We are Social Media Middleware

March 9, 2009


From Wikipedia: Middleware is computer software that connects software components or applications.

Look at the conversation prism. It attempts to round up all of the 2.0 services out there. While many more have sprung up or died off since Brian Solis remade Robert Scoble’s Social Media Starfish, the visual metaphor does a great job telling the story: There are more communication services out there than you know what to do with.

Brian Solis' Conversation Prism

Brian Solis' Conversation Prism

I am big fan of Mashable; I follow them on twitter and I read their blog daily. Their answer to managing the conversation prism…a 5 step process. I don’t think so.

Where’s the software that ties them all together? Who’s got the software stack infographic that shows how they all interoperate? Who is doing the legwork on the protocols and APIs to make sure we don’t have to think about them?

Oh, wait that’s right: No one.


The burden is on us. *We* are the social media middleware.

Maybe that’s what Eric Schmidt was trying to say when he bagged on Twitter as a poor man’s email system. Maybe he was talking about the lack of basic robustness that a “real” business needs to operate. Not that Twitter should evolve to become email, but it could certainly stand to learn from it. Kinda like the way we learn lessons from the mainframe world of 20 years ago on topics like virtualization.

As I was saying last Friday, we’re inundated with data these days and much of it is from some form of communication. Each service has it’s own interface, user experience, APIs, and clients. There are so many valuable insights we could gain from a system that ties them all together. Do I have any volunteers? It can be our ‘side project’. I’m not joking.


Both Twitter and FriendFeed have done a great job at doing the “radically open API” dance but it seems like this thought is directionally more about the aggregation/dashboard/usability thing. It’s a tough thing to hack imo because of the Facebook juggernaut; clearly they have the strongest dashboard play if they opened up the data I/O (i think in is as important as out).

That is, if you’re missing even one (or two) of the important social media services you use, the marginal utility of a dashboard is pretty small.

From Ethan Bauley on March 9th, 2009 at 6:51 pm

Thanks Ethan, I’m not sure if it is a dashboard? I suppose it could be. For example, I like Mint but Mint doesn’t really solve all my problems because it can’t get me the entire view of my finances…it’s close…but not quite there. Do you know any social media IT professionals out there that could help us with this ‘side project’?

From jascha kaykas-wolff on March 9th, 2009 at 7:35 pm

This isn’t a new situation and always occurs when there are like services that do not have interop. This is rearing it’s ugly head more prominently due to the volume of information users are consuming out of microblogs and social networks.

This has occured in other situations, like IM networks, which is typically solved first by 3rd party services, then often natively by the services themselves.

With IM networks, users were fragmented between MSN/Windows Live Messenger, Yahoo, AIM and gChat. This first gave rise to services like Meebo.com and later Yahoo and Microsoft did native interop. Native interop always has the biggest impact on users since 3rd party services obviously only benefit the very small % that know about, and adopt, that particular service. (meebo user base in tiny compared to overall AIM users for example).

We can solve this for microblogs and social networks by either (a) absorbing the interop problem as users — be the “Social middleware” as you put it, or (b) wait for interop between the services we’re talking about now — twitter, facebook, identi.ca, etc.

This definitely isn’t a new problem and we’re bound to see a glut of aggregation/interop products come on the market. Take AtomKeep for example — their core value add is just to simplify the profile updates only. Another example is Ping.fm which addresses both status update and microblogging. We’ll see more and more of these, until we get true native interop.

It won’t be long before we see someone define a standard for each, then we see uptake in adoption. How long? Your guess is as good as mine, but my prediction is that it will be orders of magnitude faster than we saw IM interop between Yahoo and Microsoft.

From Trevin Chow on March 9th, 2009 at 8:48 pm

I mumbled something on Twitter last week about wanting to move past traditonal CRM and into SRM – a social relations manager. As it is, I’m on any number of communication and relationshiop platforms. Traditional email. Twitter/Identica. Industry-specific forums. IM. FriendFeed pulls in even more from my associates. None of this ties together. I can’t see all of my messages/interactions with one person in one place.

Various CRM or other servies will let one manually tie pieces together, but I hate spending time just to build links between fragmented, siloed containers of information. I hope that the open stack will eventually lead to more seamless data exchange, but I suspect that boundaries between different information types (email, microblogging, long form blog posts, location updates, IM, etc) will remain for quite a while.

Part of the issue is one of usability. I use GMail’s web interface because I like it. Some CRM systems add some integration features, but they want to use an offline/proprietary mail client to do it, which is a showstopper. Mobile interfaces are another challenge.

It’s an interesting space. Realistically I think we’re still a couple years from seeing any major breakthroughs, but it’s a fun space to watch.

From Aaron on March 9th, 2009 at 9:39 pm

I asked this same thing a while back and all I got was, “Use FriendFeed.” I think there has to be something that we could come up with to really mashup CRM, PRM, social media activity, web activity in general and maybe even tie that into sales numbers.

I’m in!

From Neil Minetto on March 9th, 2009 at 10:52 pm

So, who starts the interop avalanche? Trevin, does it have to be a biggie like FB or can it be a consortium of smaller players?

Could a salesforce.com ever play here? Aaron are they too big and not accessible?

Neil, we have two of us :)

From jascha kaykas-wolff on March 9th, 2009 at 11:09 pm

Anyone that says that FriendFeed is a solution for interop is on some hard drugs :) It’s primarily an aggregator of your activity on other sites, and lays a thin layer on top with their notion of “like” and commenting. I have a FF profile, but barely use it.

So without the services providing some formal APIs, we’re stuck in the interim with FriendFeed-like services which need to suck in through limited APIs or even worse, parse the data.

Twitter and other services provide APIs, but not to any type of standard.

One obvious problem to overcome is the volume/quantity of information generated out of these social services. Facebook is nothing compare to Twitter, and they suffered scalability problems way before the likes of Shaq and Britney Spears joined. Their problem will only get worse.

So unfortunately, I don’t think there’s an easy answer. Services have to first agree on a standard, then figure out a way to offer APIs, and scale accordingly. If there’s no standard in place, they pass the cost of brokering data between services on everyone else.

From Trevin Chow on March 10th, 2009 at 1:10 am

You write:

“As I was saying last Friday, we’re inundated with data these days and much of it is from some form of communication. Each service has it’s own interface, user experience, APIs, and clients. There are so many valuable insights we could gain from a system that ties them all together. Do I have any volunteers? It can be our ’side project’. I’m not joking.”

As a mathematician / statistician, I’d love to work on something like this! And I actually think there are folks trying to do things of this nature. Still, the obstacles aren’t all technical. I think it may be really more about who “owns” the data than the technical challenges.

Along those lines, you might want to look at this by Nova Spivack:

Can Twitter Survive What is About to Happen to It? http://bit.ly/JA9Sv

From M. Edward (Ed) Borasky on March 16th, 2009 at 5:07 am

microblogging is really useful when you want to broadcast short updates. i am still leaning towards traditional blogging.':`

From Charlotte Hughes on May 1st, 2010 at 3:17 pm

Congrats to bit.ly and Webtrends for both making the media used socially a little easier to analyze through their announced partnership. I can’t wait to see how the integration pans out! We have been hacking with combining webtrends Pages reports (full URL) with bit.ly results, but there have been challenges. The devils in the details, but this should be exciting. Keep it up!

From Thomas Bosilevac on June 4th, 2010 at 5:10 am

Thanks Thomas. We love working with the team at Bit.ly. They are smart and doing awesome things. With our focus on innovation at Webtrends this is a really natural next step for us.

From jascha kaykas-wolff on June 4th, 2010 at 11:20 pm

[…] so many ways I’m just caught in the trap of being the social media middleware again. A few weeks ago I started to use Timely from the team at Flowtown. It has made quite an impression […]

From Marketing Iteration – Optimizing your personal publishing with Timely on February 8th, 2011 at 6:12 pm

Share Your Mind:

Fatal error: Call to undefined function aktt_sidebar_tweets() in /home/petitcou/public_html/marketingiteration/wp-content/themes/iterate/footer.php on line 2