DISQUS

Sounds like a good idea. It could fit the pre-requesite to any good ad based model, i.e., an excellent fit between content format, advertising format and user’s value proposition (more on this in this post I wrote a few months ago: http://tinyurl.com/65yms3 ).

The question is whether this scheme really fits with the core users' value proposition. Not sure.

It's just that that you can do: "help people cut through the noise to get just what they are interested in" and I would add "according to your subjective opinion of what is interesting on a given topic or what message you want to promote". Now to be very clear, it is not really a tool for "real time" news (or not yet at least) but rather for the "lost memory" and to discover content of the long tail.

Very concretely you install a plugin (only on firefox for the moment) that records your surfs whenever you want. Let's say you're doing a surf on a specific photographer (e.g., Robert Mapplethorpe), you just hit "pearl" to record and "reveal" at the end. You then will see the map of your surf in your account. You can easily re-organize/re-order this map as you wish, delete useless items, comment, etc. to build the story you want to. Basically you have crystallized your thinking on Robert Mapplethorpe. You have built an edition about him. Overtime, you will add new contents, delete some others, add new comments, etc.

Why would other people be interested on your map on Robert Mapplethorpe? Well, simply because you have cut through the "noise" of all contents on him to build a story of what you subjectively think are the most interesting aspects. And given your blog, it seems you know something about photography so you have some credibility.

How can other people benefit from your thinking? 1) You can share this map with other people via one link, embed it on your blog, etc. 2) other users interested in photography will either eventually cross your map and end up on it or 3) they will click on "pearltrees" in the upper left corner and will see the closest maps to their own map... and if they have similar interest than yours, your map will surely show up. The latter is at a map level not an account level, meaning that if you have various interests, you'll have various interest networks.

I'm of course not saying it's the only way to "scale yourself"... they are probably other ways of doing it. But it certainly is one way of doing it... though I know I'm very biased as a cofounder.... ;-)
So I'll stop it here. Just go on pearltrees.com and try it out the early alpha.

Yes fully agree on your point on integration / standardization.
On practicalities: i don't really see VCs giving 2bn to Twitter to buy Skype... ;-) But I stop on practicalities, it's not the purpose of your post, it's a sidetrack and the "merger" wording makes the point. ;-)

I understand better. Mind you by building your map that other people will cross and then explore you will deliver content (a webpage or a map of webpages) within a context (the position of the webpage within a map or the position of a map within a broader map).
Now, to be clear it is not an 100% automated process to build your map, you need to work on it: change the order, the position relative to other contents. Why? For the simple reason that the human brain will always be more powerful than any algorithm to understand a content and its context. It's the essence of the edition/curation activity. For example, an algorithm cannot replace the curator of a museum for the selection of 20 art pieces among thousands and for the order in which he/she will organize them in the museum to build a meaningful path/story about a given artist.

btw, I don't see "real time" allowing me to see more relevant content but rather to see it faster. That's what I like about "real time" but it's frustrating to lose the quality content I receive from Twitter for example... the lost memories.

You pick (again) a good pain point. I tend to agree with Fred that the end point integration is key. From that perspective merging Twitter and Skype can be attractive as it puts in one platform those 2 communications tools. But then there is still my Facebook communication, my emails (yes I still use this tool...), etc. Integrating all these requires to "liberate" data... but some players (a.k.a. Facebook) don't seem big fans of that.
In addition, fragmentation is not always bad. Different online social networks serve different type of networks (e.g., Fb for friends, LinkedIn for professional, Twitter for web people, or whatever fragmentation you want etc.). This is a translation of our offline fragmentation behavior. We have always had several social circles between which we navigate and that we keep (partially or totally) separate. What is pain is the fragmentation to "monitor" the communications from each one of them.

On a more pragmatic note, as of today it would be more Skype buying Twitter than the other way around as 1) Twitter lacks cash whereas Skype could raise cash when it does its planned IPO and 2) at least currently, Skype has a way larger valuation.

I know I’m a bit late in the discussion (actually I didn’t know you had this second blog). It’s a very important topic that you cover here and your "personalized API" post. A topic that is close to my heart.
I would summarize it in “how to access content that is of real relevance to me among today’s overflow of content?” Or in other words, “how to democratize/decentralize access to content?”.

There have been attempts. For example, content aggregators are a great improvement in that direction. But they are based on a majority logic which means it’s a “dictatorship of the majority”… and my interest may not be in line with that of the majority (not to mention that it’s more about pushing content than really sharing content).
This means that to access content of real interest to me, I need to be part of a network of people that have similar interests than mine. That was the attempt of social bookmarking. By revealing our preferences, we can access content from other people that have similar preferences. The issue is that it is based on tags… and a word depends on the context and the subjectivity of its messenger. For example, “financial crisis” or “state failure” maybe a common tag between a communist and a libertarian but it does not mean that they are interested in the same contents. So preferences can only be inferred from the underlying selected contents.
As Bonifer points out, I also think that the interconnection of the themes is very important. It allows building the context of a given content. Mind mapping methodologies are great in that sense.

Part of we do at Pearltrees is trying to solve this. First, Pearltrees enables you to easily build the maps of your web. Second, you can share these maps with friends to guide them through what you think is relevant content to them. Third, and more importantly, the maps of different users cross each other on a given web page. Pearltrees allows you to navigate from your maps to other user’s maps and thus to discover new content from like-minded people (these “crossings” are ranked by highest communalities). Meaning that collectively, the community is building the web interest graph.
We are still in early alpha (opened 2 months ago) so the product is still rough (working hard to improve it!). For english contents, points 1 and 2 are more tangible today than point 3... but it should come as US/UK members grow... ;-)

Sounds indeed a very good idea. Will try it.

Sounds interesting indeed but it seems to me though it seems to narrow very much the attention. If being an editor interests you, I can only recommend you try pearltrees... It basically allows you to be the editor of your web: build maps with web contents you select and then share these maps with your friends to guide them through the web. On top, your maps will cross other user's maps and hence allow you to discover other "editors" edition. Ok, I stop the selling piece here.

Fully agree with you. There are a lot of other arena where social networks can add value and disrupt a current sector or consumer habits.

It's going to be a major challenge to get government agencies into a customer centric culture. A lot of companies in a competitive environment don't manage to do it. So think about this for government agencies that have a natural monopoly. The issue is how does the state perceived itself towards its citizens? Is it really the emanation of the people or is it a fully different entity with a "hierarchical" view of its role?
The challenge to make is different from country to country. From my few months of living in Singapore the situation seemed much better than in most European countries. And within Europe there are some good practices that exist.

I agree with "reader" that tagging is intrinsically flawed because of the lack of context. It's unfortunate in a sense because it was a first attempt to decentralize access to content. Even at an aggregate level it does not work. Today we all use search engines, social bookmarking or voting systems to access content. However those 3 systems yield the same kind of results because they are based on a "majority" logic. However, I'm not interested by the "most popular", I'm interested by what is relevant for me, for my tastes, etc. In turn it means that today web users cannot find their ways in the mass of content available. In essence, the web needs to invent a new way to access content that is based on a logic of "proximity" to each individuals.

Very good points indeed. It's hard to see any synergies in a lot of Google's acquisition and even when they are, they often execute poorly afterward. In a lot of cases actually, Goggle acquisitions are defensive moves rather than an offensive ones.

In France the government has recently put in place several measures to help "innovation". First as mentioned by mdangear, there is a fiscal break for individuals investing in start-ups which has indeed boosted the business angel market. Second, there is a tax credit of 30% on R&D expenses applicable to all companies. Third, a start-up can ask for a special status that lowers the company social charges from 40% to roughly 15-20%. Fourth, there is a governement agency that gives zero interest rates loans for 30-50% of the initial investment required (up to a certain amount). In the latter however, the agency selects the projects in which it invests. But it basically means that theoritically 50-65% of your R&D expenses can be covered... though you still need the cash upfront since the payments are usually made after the expenses occured.

To broaden your question, a government action to put in place a given industry ecosystem (innovative or not) has been done numerous times in the past decades. Actually it was a standard policy in Europe after WWII (e.g., Airbus in France). It's also the strategy that successful developing countries have put in place (e.g., IT offshoring in India, manufacturing in China, but also smaller countries like Morocco). But there has been a lot of failures too (e.g., Latin America in the 60's and 70's). The sucess cases have in common to align (lots of) money, political will, education system, infrastructure investments, tax breaks, etc. In other words, there is no silver bullet. Moreover, the "new ecosystem" tends to build on an "adjacent" pre-existing ecosystem and culture. In any case, it's really a long term investment.

For "fixed cost" industries, you also need a critical element: "natural market" with large scale. The latter explains why, for example, the US software industry was more successful than in any European country. Building on this software industry, the web sector came very naturally as an extension.
Based on this logic and considering that the web industry is slowly becoming a "variable cost" industry, I would argue that it is today less difficult to build web start-up ecosystems in new locations (though it will be difficult to reach Silicon Valley scale).

Congrats! Happy to hear that!

It's excellent news indeed, especially knowing that Meetic launched after Match.com.
Not enough to show a change of trend but enough to demonstrate that it is possible. After all, if European companies can be leaders in other industries, there is no reason to believe they would not be able to be leaders in segments of the web industry. The latter is still very very young so there is time to see this trend happening.

I very much like #5. I don't think it needs to go through a voting system however. Actually it should not: voting systems bring up what the majority likes/thinks not what I like/think. So I would like a product that identifies people with similar interests/views than mine on a given topic and allows me to see content on this specific topic these people liked. This would be way more value-added than any voting system... Happens to be what I'm working on.

Fully agree Phil. I have to say that there is intrinsically a lot of noise in Twitter and that this "gaming" is just increasing it. I decided early on to follow only a small number of people. The implication is that I sometime unfollow people where I see I have limited common interests.

as usual great interview

Thanks for these great lists!
Since you have LeWeb up there, here are 2 other general tech conferences in European
- The Next Web conference in Amsterdam (http://2009.thenextweb.com/)
- Lift in Geneva (http://www.liftconference.com)

Extremely interesting Fred. Thanks.
It would be interesting to put these figures in perspective with the standalone impact of TV/ radio/ newspaper ads. The real power of internet ads lies when all channels are coordinated with traditional media campaigns. I had seen some research from McKinsey (sorry no link) showing that web ads had more impact early in the sales channel, e.g., during the information gathering phase.

Indeed, even if interests are not always perfectly aligned between VCs and entrepreneurs, both parties need eachother and have some similar long term interests.
IMHO, per se, the growth of secondary market is at worst neutral for GPs (at best positive because it gives liquidity to this market). It's rather the reason for this growth that is negative for GPs. However, today if I had to choose between being a PE GP or a VC GP, I'd prefer the latter by far (at least I wouldn't have all this debt to worry about).

Fair enough in this cases. I was more thinking of B2C platforms where there is no business benefits for the community. Anyway a long shot out of the box idea.

Why not giving shares of the start-up to the most active users? I see 2 main reasons for this:
1) After all without these users the start-up would be nothing
2) Its being consistent: community based products always say that the community is part of the project, that the community "co-owns" the product, that they want to build the product with their community, etc.. Then if that's really the case, the community should also get some part of the capital

Of course there are operational difficulties to do it and maybe even some legal aspects to take into account. But still could be a cool concept: a truly collaborative one.

Fully agree, nic. It reminds me your post on mobile operators a while ago.