Sunday, May 31, 2009

ORF vs BBC

BBC - the public broadcasting corporation - is widely regarded as the one broadcaster who managed to "get" new media [1] ... e.g. by opening up their archives, streaming their program (or should I say programme) , ...

Compare that to ORF - the Austrian Broadcasting Corporation, public as well, mainly funded by public license fee as well: not even close: hardly any streaming (except for FM4 alternative radio afaik) , some podcasts, no public archive...

One nice indicator of this difference is that on a BBC podcast like the Friday Night Comedy on Radio 4 they start with thanking you for downloading the podcast... whereas on ORF e.g. OE1 Digital Leben ("living digital") at the end of the podcast they tell you that "this show is also available as a podcast"... duh...

Just shows they don't even care to slightly change / cut a program in post-production for podcast... they just "copy" the radio show... instead of using the place/time on the podcast to refer to other podcasts or "real" radio program or in any other way directly address the podcast-only listeners (like myself).

Yet another chance missed.

[1] or rather "the 21st century" ;-)

Sunday, May 17, 2009

Facebook, Friendfeed, Xing and Twitter

Since Facebook became more stream-oriented there is an ongoing discussion (or rather a fight, it sometimes seems) on whether Facebook is better then Twitter or the other way round? If geeks are present then FriendFeed also comes into the discussion. So I'd like to share my thoughts on those tools, and I'll include Xing for completeness and to have a rather different platform here as well. [1]

1. Noise
One way to differentiate those platforms is noise or the concept of the signal-to-noise-ratio (for the engineers amongst us), short SNR.
As Wikipedia puts it quite well:
"In less technical terms, signal-to-noise ratio compares the level of a
desired signal (such as music) to the level of background noise. The
higher the ratio, the less obtrusive the background noise is."
In yet other terms for social network streams: "How much crap is there on a stream/feed in relation to the interesting stuff."

It of course depends on you and the platform (Facebook, Twitter, ...) what to consider noise and what to consider the signal.
Lets take the "status updates".
  • On Twitter the status updates *are* the signal... so the signal-to-noise is 1:1 or 0 dB :-)
  • On Facebook the status update are part of the whole point of Facebook, but not the only one... they are not strictly noise, but also not the only signal... SNR > 0 dB, but not too high either.
  • Xing does have status updates, but - fortunately - they are rarely used. The stream on Xing that I'm interested in is "Who changed job?" "Who got connected to whom?"... Compared to this signal, the status-update-noise is fairly low... excellent SNR there.
  • Friendfeed to me is somewhere between Twitter and Facebook.

2. Persistance
  • Twitter is only a stream. What you don't see when it floats by, you'll miss. Yes, you can search (thanks to the acquisition and integration of summize into twitter), but that's it. And as far as I saw, only a couple of weeks of history.
    So Twitter only consists of triggers, and does not really have a state.
  • Facebook has both, the stream of status updates and more static content like Group memberships, Fan-ness, etc.
  • Xing is totally static and is actually designed to keep historical data (if you wish so). On Xing I can look up which companies a person worked for in the past years. I don't have to search through status updates to find that out, it's just there on one neat little page.
  • Friendfeed like Twitter, does not really have a state, but is more browseable, again somewhere between Twitter and Facebook.

3. Relationships
  • Twitter and Friendfeed are insofar unique as they support (and are actually designed for) unidirectional relationships. I can follow you, but you still don't have to follow me. Neither technically, nor is it socially mandated. If I follow you on Friendfeed, and you still choose not to follow me there, I hold no grudge against you.
    This also means I can unfollow you anytime.
    Also, this relationship does not have a quality associated with it ("Business", "Family", "Friend").
  • Facebook and Xing require confirmation from both parties to start a relationship. Ignoring or even denying a friend-request or link there is visible to the requester and thus considered rather impolite. It is therefore also rare to unfollow ("unfriend", "unlink") someone, once the connection is established.

So those 4 social platforms have quite different qualities, and that's totally OK with me.
I use them for separate purposes.
I use them at different times.
I think I do have a different style on those.

However, I'm still not quite sure what to do with FriendFeed - a service that I really love... but whatever I do with it, I could also do with Twitter or Facebook. Currently, it's just the better and more complete userinterface into Twitter for me. I guess the main reason is, that more people/friends are on Facebook and Twitter than on FriendFeed.
So sign-up ... and follow me there.


[1] yes, I know there are others like LinkedIn, Plaxo, ... but I rarely use them, or not at all (like myspace)

Wolfram|Alpha only half-intelligent

There are numerous posts about the failure of the "computational knowledge engine" Wolfram|Alpha, just launched.

I don't consider it a failure - at least not yet.

However, one thing that I noticed: they tend to be very intelligent around your input and interpret it in the most meaningful way...

Still - they failed on an easy one.

Try to enter an ambiguous date like "4.5.1980" to find everything for May 4th, 1980... at least that's what you'd expect when you are in central europe (or more generally speaking: outside the U.S.)
Wolfram|Alpha's interpretation, however, is April 5th. ... well, that's not to uncommon for U.S.-centric sites.

What strikes me as odd, is that on the very same page, they show me the weather for Vienna, Austria for that day... nice feature... Yet, they are telling me, that they exactly know that I'm not from the U.S. and the date should be interpreted as a European date, i.e. day.month.year and not the other way round.

Guys, how could you miss that one???

Saturday, May 09, 2009

OpenOffice 3.1 is available - really ?

and ready for download.

However, after installing OOo 3.1 I could not longer work with Calc. Neither thru starting scalc.exe or opening a .ods document, or creating a new spreadsheet from within a running openoffice instance.

Through the UNO/Java API I saw that there is a problem in loading the SCMI.DLL library, but no further indication on what exactly fails there.
Removing and re-installing did not help.
Using the US English version instead of German didn't help either (could've been a build problem with the German build). No hint on google either.

Had to revert back to 3.0.1 - calc now works again.

I'm quite angry - mostly at myself, because I usually test new versions in my virtualbox... This time I did not,.... argh!!!