As I already told, I am getting into twitter and though I'm not having any significant social network on twitter, I'm starting to get some of the ideas and some others - I just don't! But what I can tell you about are some really disgusting usability issues I came across after only a couple of days of usage.

After finding out that Chris Messina is twittering (..or tweeting? What's the 'correct' verb here?), I thought that it could be worthwhile following him as we both share the interest in distributed social networks. For that Chris started the DiSo Project and I recently joined the according Google group. Following Factoryjoe actually payed off rather quickly as I received the following twitter-message:

03:35 < twitter> factoryjoe: Wow, gave my my hour long #DiSo talk in 30 minutes. Phew! #bcsd (~70 slides)

Hey, great, Chris gave another talk on DiSo and I bet that was interesting. He posted a link to slides from a former talk on the DiSo mailing list some time ago already. Those slides looked interesting to me as they were pretty clear, empty and not the usual bullet-point-shit. Especially for their appealing clearness it is hard to grasp much from them though. So - one more 'hey' again - let's use the big social networking advantage of twitter and let Chris know that we'd like to have some audio (or even video) from the talk to get deeper into it!

03:49 <   amette> @factoryjoe Same slides as in Vancouver? Is there some video (or audio) of your talk (from whereever) available?
04:08 <   amette> LOL - those @someone thingys only work when someone is following me - ok, still learning... ;)

Yep, yep, nothing to shout at twitter yet, just my own fault, I agree! But I still had to realize myself that he won't get that message and I wouldn't have realized it, if I hadn't read about that some time ago already. So after recalling this issue and accepting that I just messed up by sending useless messages out to the world, I searched for some other solution to not pollute my feed any more. Well, after STFW I found out that one can post private messages on twitter by just prepending them with "d". Ah, cool, so lets go for it:

04:09 <   amette> d factoryjoe Same slides as in Vancouver? Is there some video (or audio) of your talk (from whereever) available?
04:09 <  twitter> factoryjoe does not follow you. Send 'follow factoryjoe' to request.

Ah, well, sure, if he doesn't follow me, then it would be unfair to spam him with private messages - good point, twitter-guys!! And WTF? You guys got some command possibilities in you instant messenger interface? w00t - I love you!! I asked for just that in my previous blog post. That's just what I want!! I use my irssi with its awesomely great proxy-feature (well, nowadays just SSH ;) ) on my server for IRC and after discovering bitlbee I just ditched all desktop-clients for Jabber and ICQ. So I was right from the start very happy to see twitter integrating IMs as I therefore got one more service integrated into my (almost) ubiquitous SSH - and that even by means of the open XMPP protocol! And now I don't really need the twitter web interface at all any more, it seems. I even got a help message back that really makes sense... somehow...

... I just didn't get why I should send "follow factoryjoe" to make him follow me - it should be his choice to do so! Something like "invite" would have made sense to me and I was really alienated to issue some "follow"-command. After some more googling I gave up and just decided to try it (look at the time-stamp, it really took me eleven minutes ;) ):

04:20 <   amette> follow factoryjoe
04:20 <  twitter> You'll receive a message every time factoryjoe updates. To silence, send 'off factoryjoe'. For more commands, send 'help'.

WHAT?!? Eleven minutes searching for nothing was really the thing I was supposed to do for gaining - nothing!?! I was following factoryjoe already - so adhering to twitter's suggestion only to be told that I am now following him didn't give me much anything at all! Well, I STFW and I'm surely not thinking too highly of myself to not RTFM. So before you go hunting my ass down over not trying to get help for real, I issued the 'help' command as suggested:

04:20 <   amette> help
04:20 <  twitter> Reply with what you're doing.
04:20 <  twitter> 'invite' to invite a friend.
04:20 <  twitter> 'follow' to receive updates.
04:20 <  twitter> 'track' to track interests.
04:20 <  twitter> 'whois' for info.
04:20 <  twitter> 'off' to silence.
04:21 <   amette> help track
04:21 <   amette> track
04:21 <  twitter> You aren't tracking anything. Reply with 'track your keywords' to start.
04:21 <   amette> invite
04:21 <  twitter> Send 'invite number' to invite a friend to follow your updates on Twitter. e.g.: 'invite 212-555-2323'.
04:22 <   amette> invite factoryjoe
04:23 <   amette> whois factoryjoe
04:23 <  twitter> Mr Messina, since Jul 2006.
04:23 <  twitter> bio: As if concentrating wasn't hard enough already.
04:23 <  twitter> location: Temecula, CA, USA
04:23 <  twitter> web: http://factoryjoe.com/
04:23 <   amette> track diso
04:23 <  twitter> You'll now receive updates matching 'diso'. To stop, send 'untrack diso'.

Ok, that looked pretty nice at first, as it seemed to explain what is going on! And suddenly there's even the 'invite' command I was looking for! 'Track'ing something sounds like an interesting possibility and I decided to get some more help on that command by issuing "help track" - which sadly yielded nothing! Not even some "Fuck you! This command doesn't work like this!" - I would have actually been happy with such an answer, for real! Well, I learn fast and just issue the "track" command which then tells me what to do. I become overwhelmed with my success and issue the "invite" command because I need some more success story and really even get it! I shall say 'invite number' to that dumb bot, but I really don't know what number it means! A little bit of hubris began growing inside me and I just said 'invite factoryjoe' as one would most probably think to be a good idea as twitter treats usernames as being unique! No! That was not a good idea, it didn't help anyting and _worse_: I didn't get any feedback on that either! And it is becoming even WORSE... but more on that later or you might be suffering a heart attack right now....

I went clicking around twitter.com in my browser for some time to perhaps find a help page that I overlooked before. And yeah, for real, there it was sitting in the upper right corner: a friendly and helpful looking little link saying "help". Great! I clicked it and my page reloaded. Yes, I mean it the way I said it: my page reloaded! I tried again and my page reloaded. I logged out and clicked 'help' and my page reloaded. I used another browser, clicked help and my page reloaded. I... you get the picture. The little friendly word 'help' always links to the current page you're on. Wow, that's really helpful! Actually it would be even more helpful if they would name that link 'reload'! Another example can be found on the twitter web page. There are links to "followers_me" and "following_me" - heck, I still don't know what that's supposed to mean! Just leave the "_me" part away and it starts making sense - just please don't try to look cool for no reason.

It however occured to me that the twitter 'invite' command probably takes a phone number as argument. So I searched the web and tried to find factoryjoe's cell phone number.

04:38 <   amette> invite 1234567890
04:38 <  twitter> We sent an invite for you! Expect to hear from your friend soon.

Wow, that was easy - once you know what to do! Well, he's not my friend, but discussing this incorrect 'friend'-notion used in almost every social service is another thing. ;)

So after more than half an hour I finally accomplished what should be a simple task for a social web-service: making contact with another guy. Now quickly twittering my 'success story' and then back to writing my thesis.

04:41 <   amette> shaking my head massively about twitter usability, but need to get back to my thesis now before blogging that crappy experience I just had...
04:41 <  twitter> Oops! Your update was over 140 characters. We sent the short version to your friends (they can view the entire update on the web).
04:44 <   amette> LOL - thought it was over, but my last tweet really made me cry/laugh... expect some rant on my blog :P
04:55 <   amette> invite 23542

Nicely done, twitter folks! Giving me feedback about my message being too long and wow, you even show the full message on your webpage still. Yeah, if that would have only been true! The message on the twitter webpage is just as cropped as the one in the RSS-Feed or the one that was shown on my blog through the twitter API. This was just as blatant a lie as telling me to issue the follow command to invite someone.

And what's that "invite 23542" you wonder? Well, that was a deliberate test related to the heart attack.. all the commands I issued, which yielded nothing - they were posted to my twitter account. And even the "invite 23542", which really looks exactly like the command shall look like ("invite number"), was posted to my twitter stream. I think it's really a dumb way of exception handling to "look, if we know that number as a cell phone - otherwise post it publicly". Why don't they just use e.g. the IRC-notation for commands? Lines beginning with a slash "/" are commands, so: "/invite 23542" - "Sorry, we don't know that cell phone."

There is much more wrong with twitter usability. And it's not only about the IM integration, there's enough stuff on the web site, too! Some of it may partly be technical and related to their load issues, but still it's nothing impossible to fix and should be done fast. For example it took me more than a day to follow factoryjoe in the first place. After first clicking on "Follow" in the web-interface I didn't receive any tweets from him. I just noticed this cause I for some reason hit reload in a browser tab I had still open on his twitter page. By switching following off and on, I then managed to really be following him! AJAX is nice, but sending back success messages without really running through the logic that should do it - that's lame at least! Same goes for turning on/off device updates, sometimes it works, sometimes it doesn't. One can never be sure and actually this morning my cell phone showed me 16 SMS messages. That is some pretty nice shock every time - as a sys-admin this usually just makes me think: "Fuck! Which major system failed?!?" Well, those messages were all from twitter and I surely did not turn notification to my cell phone on. How can that happen?? And I wasn't able to switch it to my IM account through the web page. It again showed that it worked, but I kept getting messages to my cell phone. I then had to switch by sending 'off' from my cell phone and 'on' from my IM account. And apparantly it is not possible to get notified on both at the same time - why not? The list goes on and on...

... but from that IM desaster, let's learn these three easy rules by heart:

1. Name things what they are
2. Don't tell people shit that's not true
3. Give a user feedback on their actions

As you might have noticed from the right column of the blog, there is now a "currently doing" module showing my latest 'tweets'. Yeah, right, it's powered by twitter.com, which is the latest rage in the social networking sphere of the internet since some time.

Twitter is a service allowing so called micro-blogging, which basically means a blog allowing only for messages shorter than 140 characters in the case of twitter. These messages are then published on your twitter-site (e.g. http://www.twitter.com/amette and classically available via RSS (the module on the right is realized through twitters API and not RSS). Nothing special so far, except for adding a limitation to a normal blog. The interesting thing is: it is directly connected to cell phone SMS and instant messaging services! Now the limitation to 140 characters even makes sense.

Twitter advertises their service on their homepage as follows: "Twitter is a service for friends, family, and co–workers to communicate and stay connected through the exchange of quick, frequent answers to one simple question: What are you doing?" Combining this testimony with the possibility to include the tweets - as they are called - in your blog, it was a no-brainer to include it like here to create something like the Facebook status-update. One intermediate step closer to a DiSo-site. (Oh, DiSo Project, right, one thing I 'forgot' to blog about - but well, lemme just tell you that it is exactly what I have been thinking about for three years and that it will be the next big thing on the net! Not necessarily big in terms of monetization or such, but big in terms of paradigm-shift!)

As with any technology though, twitter quickly evolved differently than its founders might have thought. And I don't mean the usual 'any new technology is first used for pr0n distribution' meme. No, people are having public conversations on twitter. It's like blogs with trackbacks on amphetamines. And this actually makes my concept of using it as an activity stream on my blog pretty useless... I mean, I tried to follow Stowe Boyd for two days - and he's fuckin' badly getting on my nerves already! It's not necessarily his fault, but in the end I only receive his side of hundreds of conversations - which makes anything he says almost totally worthless! And no, I won't begin to follow his hundred followers he is communicating with to get the big picture. There's something wrong with twitter, it didn't grow with the needs of its customers. It may be a good (if not even awesome) solution for the people in one clump of the social network as they all follow each other's tweets and don't get messages ripped out of context. But that concept just doesn't scale. I might have my clump of people on twitter and we can leverage its use just as well, but for getting some insight into the thinking of other people - it just sucks! Well, on the other hand it might just as well be those people's usage of twitter... at least this time I know the spammers by name and can opt out. ;)

Another interesting service is Jaiku, which provides micro-blogging with added aggregation of activity streams from various social networking sites. This comes much closer to my idea of the usage in my blog's right column. Jaiku was bought by Google lately and since then the finnish company's (bet those guys know what a Play-Off Beard is ;) ) service is invitation and waiting list only. I signed up and now am waiting, let's see when I will get in.

In the end it's the users who decide how a service gets used, but by design one surely can influence the usage. I strongly hope for Jaiku that they don't try to be the second Twitter, they should find their own way. They have the added benefit of aggregating activity from other sites and still allow for self-typed messages. They have a good starting position for competing with FriendFeed and twitter at the same time. Google so far fucked anything social up, so I fear a bit for my finnish colleagues. On the other hand, I think twitter should emphasize on the communication and make it more easy to follow. I love to use it by Jabber, but I mostly get senseless spam - some stuff though sounds interesting. And when I see an answer to someone I don't follow, why not allow me to send an IM-message like "unfollowedGuy 3", giving me the last 3 tweets of him so I can get the big picture?

EDIT: Ok, got a FriendFeed account now as they allow it. Set up four different social sites, I have an account on, in about five minutes and got used to it quickly. No twitter-like texting, but I can include my twitter posts in my friendfeed... Dunno, if I will need Jaiku any more, if I should get in some time - Google really manages to fuck stuff up... wait? Didn't I mention something like that? :P

So, I had holidays from 2008-04-01 until this week to do some hardcore thesis writing. And as corteousness and cautiousness demand, I began growing a Play-Off Beard. ;)

One month changes a lot as it seems. Interestingly people at work had more associations with Robinson Crusoe or anti-sociality in mind. But well, I guess I shouldn't expect too much from inhabitants of a country whose ice hockey national team knows the Play-Offs mostly from TV since some decades. ;)

I got an extension of my thesis deadline until 2008-05-26 - so no shaving yet. The moustache is already becoming annoying as it feels like hanging around more in the corners of my mouth instead of above it, but it would mean bad luck to trim it. ;)

I found a beautiful Haiku in a beautiful blog post on a blog, I was almost about to delete from my blogroll. But I tell ya, this is beautiful.

Rebels have to have 
rules often to feel that there's 
a cause for their acts

found at: When Poets get angry @ Seth Godin's Blog

EDIT: Yes, surely I know that this technically is a really bad haiku - but isn't it beautiful just because of it? ;)

So apparantly nobody got, what I meant with my social networking image from the Colliding in the real world to distribute the virtual post - not even my bro as you can see from the comments to the post!!

I won't go into too much detail here - I think it's most enlightening, if one discovers it for herself. But for some clarification I want to explain it a bit at least. I already had a good VoIP chat with my bro the day after posting it, but still I can't really put it into a couple of short and good sentences.

In the first image every circle stands for a human individual. And the links are connections between those individuals, just like it is in the real world. That's what is called the social graph, it's the reality.

In the second image we see, how the so called 'social networks' like Facebook, Friendster and Xing are built. There is one central database, that connects all those individuals. To us customers this is usually sold as a possibility to utilize our social graph on the internet. This is not really wrong as those sites are capable of modeling most of the stuff that makes a social network valuable, but still it's not the social graph itself as the picture clearly shows. If you want to tell your boyfriend about your sexual desires, then you don't put that information into a database owned by someone else and tell him to pull it from there, do you? I bet you would rather have him pull it from you or even better push it to him - just like in the social graph that is the reality.

Well, the third image shows how the social graph should really be modeled on the internet. Every individual got its own node, its own database on the internet which then connects to other nodes, other databases representing other people. This kind of thing basically is possible, it represents reality by far the best and it is already being worked on. It is inevitable that it will be reality one pretty soon day!

So what about that "Social Mining"-thing? Yeah, my bro already said it in the comment to the last post - this thing is not completely new. When you put 'social mining' in parentheses on Google (and even better Yahoo), then you get lots of hits telling you about it. Well, still I take that as a strong indicator for it not being anything mainstream yet. If it would be, then one would get lots of good hits without the parentheses. And the current hits mostly talk about gathering information from your Outlook contacts and usage. But basically, yeah, that's it! The thing I was ponding here with "Social Mining" was more: When this model of decentralized social networks becomes reality, then the most valuable player in the marketing field will be the one that can produce the most valuable information from the data available on the decentratlized network. You can't just go Data Mining in your Facebook-Database and tell companies what kind of book this guy bought and the reason for it is most likely that he saw a similar book on three of his friends 'I bought'-lists. Nope, with a decentralized network every individual owns her own data and so this will become much more difficult. And those special Data Mining skills will be known as Social Mining imho.