Mistakes in Proprietary Social Networks

In the early 1990s, the telcos were racing each other in building their proprietary networks.  Metcalfe’s Law and the power of networks was starting to be understood, and the quest to create the dominant network was on, with the "networks2 intelligence" positioned as the key differentiator.  In France, we saw a version of this with the Minitel.

Then came the stupidest network of all: the Internet.  All it knew was to push IP packets around.  There was no intelligence built in to the network.  The intelligence resided on the edge.

In Turkey last week, we saw the launch of two proprietary social networks by two of the three GSM operators.  Turkcell created Turkcell-im Benim (translates to My Own Turkcell), and Avea came up with Patlican (a play on words, meaning both "eggplant", and "you will explode").

Turkey already enjoys a healthy set of social networking platform, ranging from Yonja (which started out as a Friendster look-alike) to Mondus (which we launched as a social utility/platform 3 months ago.)  Here comes a telco powered platform. And by the way, you can really interact only with other customers of that GSM network.  Quite a uniting characterstic, is it not?  I really feel like I have a lot in common with other Avea customers.

The telcos should treat the independent social platforms as the stupid network, and try to bring their intelligence layer on top of those, instead of trying to build their intelligent networks.

10 thoughts on “Mistakes in Proprietary Social Networks

  1. Patlican negetirecek/götürecek?

    Yazmassam patlicam.
    Önce kasadan ne kadar para götürdü/götürecek bir bakalım:
    Tasarım rahatsız edici, çok renkli ve estetik değil. Trendlerden uzak. Bu eleştirilecek bir şey olmayabilirdi, eğer Avea trendleri takip eden gençler için t…


  2. I wonder if this is similar to alumni organizations that are very focused on their closed, proprietary networks online. They risk ignoring the independent, third-party platforms that (though stupid in some sense) comprise individuals who connect each network’s members to other networks in turn. Instead of “pushing IP packets around” they are pushing connections between members (nodes). This lets the overlapping external networks behave as if they were smarter than they really are!


  3. Patlican – Sermaye Yoksunu Pazarlar

    Cem Sertoğlunun son blogu Türk web endüstrisindeki ciddi bir sorunu su yüzüne çıkartıyor. Sermaye kıtlığından dolayı, sermaye aşırı değerleniyor, ve iş gücü değerini yitiriyor. Tipik bir Ekonomi 101 arz-talep dengesi sonucu….


  4. Could you define what you mean by intelligence layer, and what you mean by stupid layer, as both of those can really be stretched to mean different technical implementations? One cannot blame the telcos for UU land-grab when there is so much copy-paste innovation going on (http://www.techcrunch.com/2007/05/14/web-2-in-germany-copy-paste-innovation-or-more/). After all what makes Yonja more than a localized Friendster? By the way, is Mondus a localized Facebook?


  5. @Andrew-
    Interesting analogy for the human network.
    What I mean is that networks should be just that, nodes and connections, and should act as a platform for applications, where the intelligence (or rules) reside. In a technology sense, the network is the Internet. In a people sense, we don’t know what the network will be. The identity/presence rlated “stupid network”, in my opinion, has yet to emerge. Facebook is making a strong push, though. Yonja, I think, is now a major Turkish medium which has grown with the FOAF model. Mondus, strongly inspired by Facebook, is venturing to create social layer for Turkish internet.
    If you’d care to elaborate more, I’d attempt a reply.


  6. That makes sense. In execution however, especially for a financially constrained start-up, it gets pretty difficult to separate the backend from the front UX. For example, for Mondus to be truly “stupid,” it would have to go into the real nuts and bots of the platform (such as Facebook’s APIs, Microsoft’s ABCH, or AffinityEngine’s turn-key solutions), eschew the smart facade, and private-label its offering to the telcos. It is hard to do that sell when you have to consider a run-rate and some aspiration for a PoC; and that PoC ends up being the most common applications built on top of the stupid layer. A lot of players therefore go from “building the layer” positions to vertical specialists. A lot of great coffee topics here. I am disappointed that my Istanbul trip got delayed yet again to this winter, as I have to go to Asia for business now. I’d love to meet face-to-face someday and get into it 😉 Very excited to see some of these developments back in the ‘anavatan’


  7. My registration attempt to patlican was a complete failure. They sent my cellphone an short message containing a password which i submitted to the registration form. I was then taken to another form with registration details. No restrictions for the password format were mentioned, so i chose a 4-digit one. Then i happened to realize that it should have contained at least 6 characters. Now the process won’t end because i keep getting an annoying “please fill out all required fields” although i have done so.
    I wonder how those guys can launch and promote such a huge project before completing a proper test process. Although i couldn’t manage to register the website, i’m afraid they’ll still be sending unsolicited short messages as no membership agreement exists to protect my privacy.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s