May 24 2010

I’m Dreaming of a Better Social Media Client

Published by at 8:28 pm under Technology   

I’m not a big social media guy. I’m certaintly not a social media consultant, nor a maven. I never used MySpace at all, and I was not among the first to use Facebook, LinkedIn, Twitter, etc. But I do find all of those useful to keep in touch with a bunch of people using all of the above, and I’ve grown quite frustrated with the sorry state of the client applications I’ve tried. Even those whose features work well and look good, don’t really go after the core problem we all either have it or will hit: information overload.

Here is what I really want in a social media client application for “power users” who receive a lot on their feeds: follow a lot of people on Twitter, have a lot of friends on Facebook, 500+ on LinkedIn, etc. Today, these are power users. Over the next few of years, this will be “everybody”. Most of these features make a lot of sense for a business managing its presense.

Table Stakes – The Basics

Support the Big Three (Twitter, LinkedIn, Facebook)

… and hopefully several more. But don’t even come to the party without the big three. I’m looking at you, Twitterific on the iPad, which I otherwise enjoy (and use every day, and paid for).

Ideally, RSS feeds would also flow in, and perhaps email and SMS too. But I don’t want this to be a “unified inbox” to replace an email client; this information would appear here as context for smart reading.

Run On Many Platforms

Mac, PC, iPhone, iPad, Android, Linux, maybe even BlackBerry. It’s not necessary to start with all these, but the target should to end up with all of them and more, with the core features present everywhere. I’m not looking for crappy ports though. Native, good citizens.

Keep Track of What I’ve Seen

Keep track of what I’ve seen, automatically. Don’t show me again unless I ask. But the act of closing the app should be meaningless, in that it should not mark all data as seen. An example of what not to do is TweetDeck, which has various settings for this, of which I can’t find any combination that does the Right Thing.

Next, the less common ideas:

I Paid for a Lot of Pixels – Use Them

Single-column feed display GUIs? Great idea for a phone. Silly on a PC.

Like most PC users, I have a wide, high resolution screen. Like many power users, I have two screens on some computers. I payed good money for all these pixels because I want to use them. Therefore, when I’m trying to catch up with all these data/tweet/etc. feeds, I want software that makes good use of those pixels. Show me a rich, dense screenful of information at one. Make it look like a stock trader’s screen (or screens).

Our Eyes are All Different – Give Me Knobs

I don’t want extensive customization. I don’t want a whole slew of adjustments. I don’t want a Preferences dialog with 82 tabs. I don’t even want themes. I want a good, clean, default design… but with a few well-considered knobs. Perhaps something like so:

  • font/size knob – because my eyes might work a bit better or worse than yours, and my screen might be higher or lower resolution than yours.
  • information density knob – because sometimes I want to admire a beautiful well-spaced layout, and something I just want to pack more information on there.

Aggregate Across Networks

Many of the people I follow, post the same data to at least three social media outlets; then a bunch of other copy/paste or retweet it. Please stop showing me all that duplication!

Instead, aggregate it all together, like Google News does for news sites. Show me each core message once, and then show a (dense, appropriate) display of who/how the information came in. Include a sparkline and other charts to show the continued re-arrival of that same data. This way, I won’t have to endure the duplication directly, and I can more clearly see how information traverses the (underlying, human) social network.

Some Tweets are More Equal than Others

In an ideal world, every Facebook update, every Tweet, would be a precious flower, to be admired in depth. We don’t live there. Instead, there is a lot of noise; an example fresh in my mind as I write this is the TV show Lost. It may be a great show, but it’s not one I watch, so to me all the Lost chatter is noise. I’ve probably scanned/scrolled past a couple hundred of them (some of them duplicates) over the last few days.

Therefore, a good social media client will make it trivial (one click) for me to tell it which bits I am interested in and which I’m not. I’m not talking about a scoring system, just a simple up/down arrow, for a total of three bins:

  • Important
  • Bulk / default
  • Junk

Apply some automatic classification mechanism (like the naive Bayensian that’s been common for several years now in email spam filtering) to learn from my votes and apply those to future data. By default, highlight the Important, show the Bulk, and hide the Junk.

I Have Several Devices – Sync Them Now

I might look at this river of news on my Mac in the morning, then on my iPad at lunch, then on my Linux netbook in the evening, then sneak an iPhone peek at bedtime. Keep all that “what I’ve seen” and “what’s important” data in sync across them. This means that my dream social media client needs a backend service behind it. It is not necessary for the data feeds to flow through the backend system (thought it might be useful); just the user’s attention metadata.


I believe that most or all of those features will be common in a few years. But I’m annoyed by the tsunami of social media feeds now. Is something like this out there? Where?

I could build such an application (with some help!). I’ve worked with APIs of all flavors. I’ve done mobile. I’ve created GUIs that elicit a “Wow”. I understand servers, and¬†asynchronous¬†operations, and scalability, and SaaS. But if I built it, would anyone *buy* it?

If you found this post useful, please link to it from your web site, mention it online, or mention it to a colleague.

4 responses so far

4 Responses to “I’m Dreaming of a Better Social Media Client”

  1. Well, I would pay — totally agree with what you’re saying! I’m also astonished that good filtering and cross-device sync isn’t already standard.

    But on the other hand, there are lots of “unified inbox” startups, and it’s a tricky business to be in — people have to stop using what they have now and start using your app. It requires a behaviour change, which is difficult. And it’s hard to enter with a minimum viable product, because everyone will expect your app to already be very polished.

    I’m sure it will come, but it won’t be easy.

  2. Konrad Garus says:

    Needless to say, it depends on price. More likely on the iMarket or mobiles. :-)

    I would consider it even if I only have two Linux boxes – for the synchronization, ignored topics and scalable UI.

    It seems that a web front-end could make it a lot more accessible.

  3. Kyle Cordes says:

    Good point Martin – I added a bit to clarify that I don’t have in mind an email-client-replacing Universal Inbox. More like a universal social media consumption, visualization, and monitoring tool.

  4. Jeff Barczewski says:

    Kyle,

    I agree a classification mechanism would be really needed as the amount of data continues to grow. And this learned data should be usable from all your clients, including the things you have read and any other common settings or data.

    Also good filtering for facebook apps, I mean who cares if “you found the golden goose and got 50000 points.” :-)

    I think the features you mention are good. As usage of these networks continues to grow, you need smarter clients to help you find the info you are interested in, seamlessly across everywhere you read it. We’ll be needing something an order of magnitude smarter than the apps currently out there. So we need to think outside the box.

    I think it would be possible to build something like this with HTML5, JQuery, and a server side REST web app which holds the user data and does any complex computations. That would also give you decent portability and make it easier to distribute and keep updated.

    As for whether users would pay, one indicator would be to see what social apps users pay for now. This would be a starting point, then obviously if your app has killer features, it could get more of the market.

    For some additional research, here are some articles that I found useful.

    This article has some screenshots, feature lists, and requested features
    http://www.snipe.net/2009/05/comparison-of-desktop-twitter-clients/

    This comparison has screenshots and also a comparison table for seeing features by app
    http://www.omgubuntu.co.uk/2010/02/twitter-applications-for-linux-ubuntu.html

    Jeff