DISQUS

Noise has always been the problem for information discovery, email lost the battle because of spam and although RSS readers provided a reliable opt-in solution it doesn't scale as much as we would like to (I used to rely on Google Desktop for alerts and Google Reader for delayed information discovery).

Twitter brings social filtering and although there is already noise, in this case I don't need to be subscribed to the publisher to get the news I am interested in (Twitter clients have replaced my use of Google Desktop for alerts) and if I still have time to get more news I go to FriendFeed to check what has been shared or commented. Both are the best tools available currently until track filtering arrives.

Of course, if you don't have too many feeds or you don't need close to realtime capabilities to receive the news, RSS readers are still good enough.

Watched the "conversation" yesterday and it was quite useful to understand where Twitter is right now and love the summary, helpful to refer to those who don't understand Qik social live streaming value and just want the facts.

Looks like at the beginning, Twitter was really just a microblogging platform and the problems with integrating the whole potential a messaging protocol (XMPP) can provide to the infrastructure are the architectural problems they are facing. I was surprised that they mentioned they only have one copy of the data in the disk with multiple caching updates in memory, they really have pushed the limits with this architecture and the move to a messaging platform it's going to help the downtime, hope it doesn't take them too long.

There is a opportunity for the appearance of a WordPress in microblogging (maybe a FriendFeed or Grazr open source). And for messaging, the fact that FriendFeed looks email as their competition (i.e. rooms=RSS mailing lists) signals they have already thought beyond microblogging. If more applications are built related to the mobile web like Moblf.com, then Robert is right and Twitter loyal users can look FF as a viable option.

With Nick's explanation on why de-normalization is needed to scale, it is clear one of the complex issues Twitter has to solve. The other one mentioned in this post is the business model, or when to charge.

Angus has the right idea - charging the followers - although I don't agree on the analysis. Still using Robert as the super-user, he should not be charged because of his tweets, but for the number of people he is following. Each tweet sent by the friends Robert is following will be copied on his queue (well the tweet ID) and the size and freshness of this queue (visually the 'With Others') can be used as the factor to charge.

Robert,

Have you read the article referred by Al3x in Twitter's dev blog: http://www.hueniverse.com/hueniverse/2008/03/on...

It explains why there is a need to duplicate copies, the main reason is the speed for other APIs (i.e. FriendFeed) to read them and one of the reasons why there is not a huge delay (i.e. 10 seconds) when the tweets appear in FriendFeed, without the copying that time will be longer.

As Matt says, sometimes free is enough. In the free world there is always room for improvement, and that is where a good idea becomes sustainable. Hank may have a product that users may pay for, but he needs his "free" enthusiastic users to drive improvements that will justify charging for the product.

Great news and looking forward to the post announcement on Monday, +1 to the request to have a RSS feed. I will definitely go to Cnet TV to watch the show if there is not a feed but it will be difficult to recommend to friends if none is available.

From the article: "Microsoft seemingly barred from striking pacts with distributors, only one avenue appears open to it: paying more friendly visits to its Fortune 500 customers, seeking direct licenses."

Microsoft - I hope - knows it is not wise to start the patent armageddon using law suits, because there is too much to figure out yet. 235 patents were identified and they are working on strategies to use that knowledge to get IP royalties but it is still a work in progress IMHO.

Microsoft is starting to understand open source so they are trying to learn by doing but free software is still so difficult to understand for them and this is where they are testing general reactions with this type of interviews.

This phonetic pattern search approach looks really promising, the index speed is extraordinary and it looks like Nexidia knows where their added value is: content. It is multilingual - due to the phoneme analysis - so this can have huge impact.