Lars Trieloff's Collaboration Weblog

Now that I have the attention (uppmärksamhet) of my 15 readers from Stockholm - I would like to inform you that Peter Svensson is organizing a Web Monday in Stockholm. Web Monday is a tradition of the german Web 2.0 community, a

Web Monday is a decentrally organized, informal meetup focused on Web 2.0 in the broadest of all definitions, that wants to connect users, developers, founders, entrepreneurs, scientists, web-pioneers, blogger, podcaster, designer and anyone else interested. Goal of Web Monday is one the one hand a better connection of the Web 2.0 crowd in Germany and on the other hand a deeping of the cross-atlantic exchange of ideas

(from the Web Montag Wiki) If you are interested in taking part in the first Web Monday in Stockholm, browse to the Wiki page, add your name to the list of participants and perhaps your presentation to the list of presentations and demos. In a regular Web Monday there are up to four presentations, everything that does not fit in this schedule will be covered on the next event.

Interestingly, the organization of the Web Montag is a good example of Web 2.0 communication. Peter picked up the idea from my twitter feed - and I got his announcement via twitter as well. I think this is an awesome idea and if you happen to be in Stockholm at Feb 05, 2008 (yes, it is a tuesday), you should consider being at the first Web Monday in Stockholm.

In my daily RSS skimming session, two articles managed to stick: Andrew Gent's Web 2.0 and the Lack of Process and Michael Marth's Data First ist gaining traction in the industry. While Andrew criticizes that Web 2.0 technology comes without a process and without structure, opposed to traditional tools in the enterprise such as Content Management Systems (CMS):

By that I mean that the technology itself makes no assumption about how or why the technology would be used. What is the usage model for Twitter? Who should blog? What should you use a wiki for? The answer -- if you bother to ask -- is usually "whatever you want!"

Interestingly Andrew opposes these "process-less technologies", also including blogs not only to CMS, but also to E-Mail, Instant Messaging, IRC and web pages:

Email, for instance. Process is designed into the very core of most of these older technologies. You have an email client. You choose who to send email to. They can read it, reply to it, forward it, save it, and delete it. That's all. The technology embodies the processes previously defined for physical mail.

If this is an example for the process that the tool makes assumptions about, then there is definitely a process for Wikis:

• You read a wiki page
• You edit a wiki page
• Other people read your contribution
• Other people edit the wiki page
• Repeat

There is a process for blogs:

• You write a blog post
• Other people read your blog post
• Other people comment your blog post

Out of this process emerged new technologies that were intended to improve the process. In the blogging space examples are Trackback and CoComment, which improve the process of commenting in a blog. Even twitter has a process, or an etiquette: "Follow your followers, to allow discussions" is one aspect of this etiquette.

And even tools like E-Mail, Instant Messaging and IRC needed some time to develop their own etiquette: How much time am I allowed to let pass before I answer an E-Mail? Should I send instant messages to people with an "away" presence state? It is not the tools that create the process, its the people. People decide how to use Twitter, how to use Blogs, how to use Wikis, how to use E-Mail, how to use Social Networks, how to use Instant Messaging.

People first, process later

If you try to implement technologies the other way round, without either implementing or improving existing processes (which classical enterprise software does), or allowing people to define and develop their own processes (which Enterprise 2.0 does, but the same applies to E-Mail, Instant Messaging, etc), you will fail, because your software is not able to meet the needs of your users.

A similar approach, in a different domain, is described by the keywords data first, structure later. In software development, we are obsessed with structure. Having ontologies, taxonomies, schemas, class hierarchies, typologies, classifications, structured data seems the way to build successful applications. But as in the enterprise, too much structure without relation to the reality (here people, there data) will render your classification useless. Therefore a new class of frameworks is emerging that does not enforce structured data, but allows structurable data. If you want to know what this means in terms of technology, have another look at Data first is gaining traction in the industry.

Yesterday, Dopplr, a social network that allows travelers to track their and their friends travel plans and shows friends that are traveling to the same city you are going to, has launched yesterday and allows registration without invitation. Dopplr is one of my favorite social networks and I am a happy user of the beta for two months now. So if you are traveling from time to time and want to know if we happen to be in the same city, join Dopplr and share your trips with me. I am happy to share as well.

As my readers know, I am aggregating my social network activity in a sidebar of this weblog. Today I am introducing a new feature: The Social Activity Index.

As a blogger I am blogging with varying frequency. There are times when I am full of ideas, blog and bookmark at lot of things, and there are times when hundreds of unread e-mails stack my inbox and I am unable to post a single bookmark to del.icio.us. Based on the data I already collect for the activity aggregation, I decided to boil down all these activity data into a single number by weighting time, type and frequency of an activity and combining this in a single value.

Before I published this post, my activity index was 35, now directly after publication it will be higher, but unless more activity follows, this number will fall again.

For you this number is an indicator of my activity and creativity. If this number is high, I am likely to be open to new ideas and to respond to any new input very fast, if this number is low, you can expect to wait some time for an answer.

What is a high and what is a low value for the activity index? I do not know yet. Let's monitor it for a while and find out.

Andrew McAfee writes about the concetric circles for knowledge workers - there are close colleagues, people you directly work with, people you could potentially work with, other people in your company and people you have no relation to. Social networks are tools that allow you to widen the second circle. They allow you to get to know more people for possible collaboration you do not know already and they allow you to manage loosely coupled relationships to more people than you could without tool support.

A recent piece by Alex Iskold: The Social Enterprise - What Works, and What Doesn't says social tools in the enterprise allow communication paths outside the organization chart graph. The reason is that classical tools for enterprise organization like the org chart focus on positions, not people. To illustrate this, you could use a sketch like this: We can see clearly the boxes and arrows, indicating the organizational structure. As for the people, the organizational contents, we can only see the head clearly, for all other people in the organization that we do not know directly, vision becomes blurry.

With social networking tools it becomes possible to see the actual people behind the positions, you can find out about interests, activities and create an image of a person you have not yet met in real-life, but which could help you in a future challenge.

This allows you to create connections and social structures that span multiple branches of the organizational tree. As an effect, people become more important than their position in the org chart and you can create more decentralized organizational structures like virtual teams.

As a side-note: While writing this post I realized how this people (social content) over organization (social structure) notion reflect David Nüscheler's mantra of "content first, structure later" applied to social systems.

To my suprise I found out that someone posted my presentation slides on slideshare.net before I actually held the presentation. I usually post my slides only after the presentation to update them to reflect questions that came up in the Q&A part of the session, but at the Web 2.0 Expo in Berlin things seem to work a little bit differently.

The presentation went very well and we had a very interesting discussion at the end, so I would guess the awareness of the importance of internationalization in Web 2.0 is rising. One of the most interesting aspects brought up was: Is providing textual translations of a web application enough?. My take on this question is that providing a multi-language user-interface is only the first step in true internationalization, afterwards you have to go for localized content, design and business processes that respects the culture of your audience.

If you are at my presentation, please make sure to give feedback at feedback.berlin.web20expo.com. Thank you.

Believe it or not, for more than one year I have an article draft saved in my weblog system, titled "Social Software". The original reason for writing this was that Dan Diephouse wrote about O'Reilly's Connection, a now defunct social network for geeks. Some days ago Brett Porter wrote about his experiences with social networks, and I decided to start over with this blog post.

Being user of some social networks of different flavors, I noted that the number of connections I have in different network differs strongly. One reason is that some network have a larger user base, others make it very hard to add connections, others imply a closeness of relationships that I do not want to share with everyone. So, what level of social relationships does being connected in one network imply?

Facebook

Nothing. It basically means "Hey you are on facebook as well!"

Xing, Plaxo Pulse and LinkedIn

You might be useful for my career, I better keep you in this network.

twitter, Last.fm, Flickr

What you do (listen to, or photograph) is interesting for me

del.icio.us and Digg

What you read and think is interesting for me

Dopplr and Plazes

If you happen to be in the same place with me, it would be nice to know so we can meet.

The most valuable network for me (apart from the address book synch of Plaxo) is del.icio.us, because it allows me to track and share ideas with people I find interesting and this is despite the poor network capabilities of del.icio.us. So if you would like to connect with me in del.icio.us, just leave a comment with your username,

A browser crash has killed my first attempt to write this entry, perhaps just to make me write a more dramatic first sentence. My prediction is that 2008 will be marked by collaboration convergence, this means

• tightening and deepening the connections between collaborative applications and related fields like social networking, content management systems, enterprise resource planning systems, business intelligence solutions, business process management and so on.
• blurring of borders of individual collaboration tools like digital asset management systems and wikis, wikis and blogs, blogs and forums, forums and mailing lists, mailing lists and chat, chat and instant messaging, instant messaging and twitter, twitter and time tracking, time tracking and task management, task management and workflow management, workflow management and document tracking, document tracking and document sharing, document sharing and digital asset management
• deeper integration of web-based collaboration software and desktop-bound productivity software. Right now, some people equate web-based productivity software that allows a degree of content sharing with collaboration software, but this will change.

Signs of this integration are:

• EditGrid's plugin for Microsoft Excel that allows users to edit spreadsheets on the web in EditGrid that are automatically pushed to the opened Excel window combining power of desktop tools and ubiquity of web-based tools, both to collaborate.
• Google's integration of desktop widgets that come with Google Desktop Search and iGoogle, allowing your personal web-page to embed real-time information of your hardware state.
• Mozilla's WebRunner that allows you to install web-applications like GMail or Google Reader as standalone desktop applications (via Marcel Weiss)
• Blogbridge, the feed reader as an example of desktop software that is truly web-aware.
• Apple's .Mac, back from hibernation combines desktop remote-control, a virtual hard drive, a photo gallery, and a web/IMAP mailer that is tightly integrated with the desktop operating system

Many people have been criticizing Apple's strategy of tight coupling of iLife to .Mac, but this coupling - publishing photos right from the desktop application (iPhoto) to the web and re-aggregating photos back in the desktop application is a good example of the future of convergence.

The german startup Spielerkabine.net, that a friend of mine joined recently just had to change it name from Mannschaftskabine.net, because somebody used the publicity they gained during their pre-launch to grab the trade mark "Mannschaftskabine.net" just three days after they made the first press release. The lesson for everyone considering to launch a startup is: claim not only your domain name early, but also register a trade mark as one of the first steps.

More coverage on this story (mostly in german, though)

• Packt die Taschentücher aus – es gibt Tränen wegzuwischen
• Mannschaftskabine.net heißt jetzt Spielerkabine.net
• Web 2.0 Abzocker
• Aus Manschaftskabine wird Spielerkabine

I wish you all the best luck with the new name, and keep up the good work!

Javascript seems to be the hottest language on the web these days. Not on the client side, but on the server side. Today I came across Phobos, a server-side Javascript framework with good integration into the Netbeans IDE.

Some other interesting frameworks with server-side Javascript support are:

• Helma
• TrimPath Junction
• Rhino on Rails - it is not a framework, but the pre-anouncement of a framework in development
• Butterfly
• and of course: Cocoon, my personal favorite

The advantage of server-side Javascript frameworks: You can use one language for much of the web programming on the client and on the server-side, Javascript is easy to learn, yet powerful and incredibly flexible and as most of these frameworks are powered by Rhino, you have access to all libraries running in the JVM, and that is a lot.

Webmasters and bloggers that want to publicise their content often add links at the footer of their website that allow posting a link to a social bookmarking site like del.icio.us, digg.com, reddit, ma.gnolia and others. But with the increasing number of social bookmarking services in use, and webmasters that want to support all of them, those litte icons are slowly taking over your website. (See David Trowbridge's blog for an example).

But there is relief: With Share This Link bloggers and webmasters get the opportunity to support many social bookmarking services at once by linking to Share This Link which is a meta-social-bookmarking service that forwards the bookmarker to her favorite service.

Share this link is heavily inspired by Alex King's Share This Wordpress Plugin, uses the icons provided by the Share Icon Project and was written using Apache Cocoon in one afternoon.

So, it is time to act now:

1. Share the Share This Link page with your favorite social bookmarking service
2. Clean up your blog template with Share This Link

Marjolein Hoekstra writes about Power 150, an impressive example for the power of mashups. This web application is a search engine that is based on the 150 most influential marketing weblogs and contains everything you need for blog-based marketing research with a small number of hand-picked sources.

In her weblog she describes how to orchestrate OPML, RSS and Grazr to accomplish this mashup.

Alex Faaborg writes about icon sprawl resulting from the increasing number of applications and web services that support structured web information like feeds (RSS and ATOM), reading lists (OPML), events, contacts and locations (Microformats): Structured Data Chaos. For every combination of structured data and client applications for structured data, there has to be an icon or action that adds visual clutter to the web user interface.

A similar problem appears when considering social bookmarking services like Digg, del.icio.us, reddit, mister-wong.de (a popular german social bookmarking website) and others. Wired News calls this: Battle Over 'Iconistan' due to the increasing number of bookmarking icons that clutter the interface of news sites and weblogs.

Alex Faaborg suggests to solve this problem by implementing a solution in the browser, similar to what has happend with feed support in web browsers: The web browser is able to autodetect the structured information and cares for integrating the client application.

I've got two questions: Will browser development keep pace with the development of structured web formats or will the world have to wait another two years for an update of the world's most widely distributed web browser? And so far I have not seen any proper integration of web-based feed reading services that is supported by the feed autodiscovery feature of web browsers.

One commenter to the Wired News post recommends a wordpress plugin that hides all social bookmarking icons until you click a "share" button, something like a meta-social bookmarking plugin. It could be a solution to have a similar meta-web-service (in the web 2.0 sense, not in the SOA sense) that is able to push structured data to the desired web-based or desktop client.

Nearly everyone that knows the terms Web 2.0 knows the "Web 2.0 definition game": Ask two people what "Web 2.0" means and you will get 3 definitions. 37signals asked 500 people and got 500 definitions and while reading those comments is quite entertaining, I was asking myself "What are the words used in most Web 2.0 definitions". I feed a word frequency analyzer with the results, removed obvious stopwords and got following list:

1. Web (320 mentions)
2. Applications (and Apps, 128 mentions)
3. More (99 mentions)
4. Ajax (98 mentions)
5. New (75 mentions)
6. 2.0 (59 mentions)
7. User (51 mentions)
8. Me (47 mentions)
9. You (39 mentions)
10. Desktop (35 mentions)

runner-ups are: Internet, People, Sites, Content, Think, Using, Software, Interactive, Next and Services.

Now, when you build your own defintions, remember to use the Top-10 words for defining web 2.0 to stay folksonomy compatible.

Web 2.0 means more new ajax applications for desktop users like you and me.

If you've got other ideas, add them in the comments. Interestingly, words like Social (23 mentions), Hype (21 mentions) or Collaborations (13 mentions) which had been part of my Web 2.0 definition don't show up in the Top-10 and Folksonomy is mentioned only once and Tag and Tagging are mentioned ten times.

The concept of tagging is not new (in the old times it was called keywords), but with web applications like del.icio.us (which I use regularly), Technorati or flickr, using tags or keywords as lightweight metadata has become popular again. The nice thing about tagging is that it allows you to categorize data according to more than one criteria, opposed to strictly hierarchical organization schemes like folder hierarchies or taxonomies. What is new about tagging in Web 2.0 applications? First, there is the concept of folksonomies which helps uses finding the best tags based on tags assigned to an item by other users. But the most important improvements are new user interfaces that make tagging very easy by adding type-ahead suggestions, browser extensions and bookmarklets and tag clouds that make tagging an web item just a matter of klicking the bookmarklet, klicking some suggested tags and hitting return.

The exciting news is that tagging is slowly moving from the web to the desktop. Desktop innovation happens at slower pace than web innovation, mostly due to the fact that there are much longer release cycles, but some applications like leaftag for the Gnome desktop (screenshots, video), the Quicksilver tagging module for Mac OS X (described by Livehacker as 'Metadata as a filing system') and some interesting fake screenshots of a tagging feature for Mac OS X Leopard, which look very good show the direction of development. All modern filesystems support metadata as file attributes. It is now up to the desktop developers to implement tagging interfaces for this new kind of lightweight metadata.

If we take a look ahead we will see in some years metadata-enabled desktops, metadata-enabled websites and tagging an accepted orgainizational principle. What we should be looking for is a metadat-enabled workspace environment that takes up the opportunities of tagging and lightweight metadata and support sharing of this metadata to be able to organize the data of collaborative workspaces, knowledgespaces, taskspaces and conversationspaces. An example of a tagging-centric organization of conversationspaces is vennt, an online forum software I discussed before.

last.fm screenshot

Last:fm's whole website has been overhauled to a new design a like very much. Additionally the Last.fm Player has been replaced by a new Last.fm application that offers a much better interface to tagged music and your personal profile.

Last.fm is an online service that tracks your music listening habits and compares your musical taste to the taste of others. Based on the taste of people with very similar listening habits (your musical neighborhood), you get music recommendations. Finally there is a radio service that allows you to tune in to your personal radio, based on your musical taste, your neighbor radio, your recommendation radio or specific radio stations for tagged music.

I've got an invite to peopleaggregator.net. Raju Bitter has already told me about it and recently blogged about the opening of the peopleaggregator beta. From what I've head and read, it should be the best thing since sliced bread, so I immedeately registered after I've got my invitatation mail.

But I was, to put it mildly, surprised about what I could find inside: It is a combination of Myspace-like profile pages, friends network, personal blog, bookmarks and some other services, glued together by microformats and structured blogging. I do not really see the point of this application. Nearly all users already have a weblog, so this weblog feature is not neccessary, the bookmarks feature is duplicated by del.icio.us and so on.

What I would expect from a people aggregator is an aggregation of existing sources of information (blogs, bookmarks, photos, mailing lists) and a connection with friends that works across social networks, e.g. by importing my last.fm, openbc.com and linkedin.com contacts. What is the benefit of another Myspace clone with geek-only-features like microformats and structured blogging?

Buy the way - the peopleaggregator designers seem to have taken the article The MySpace problem to heart.

Do you really belive this web 2.0 thing is about creating web-based clones of desktop applications? I have seen thousands of web applications with calendar support that all try to implement a calendar interface. But only few succeed. One example is Google Calendar which is quite well-made, but does not yet reach the ease of use of a desktop calendar application like Outlook, iCal or Evolution.

Why are web-based calendars nonetheless popular? The most important reason is that they allow exchange and sharing of data is a very easy and transparent way. No special client software needs to be installed and any user can publish and subscribe events.

This approach of building your own calendar interface in order to foster sharing will soon come to an end because in the future every user will have an desktop calendar interface that allows publishing and subscription of events. The feature to be able to subscribe to your friend's or colleague's events or important events of sports teams, public holidays has been made available to a wider audience by Apple with their iCal product which allows subscribing to calendars via HTTP and publishing via WebDAV, which is, essentially HTTP, too.

Recent versions of Novell's Evolution support the subscription feature as well and being able to publish and subscribe to calendars is one killer feature of Google Calendar. The latest and most important step in this series of development is the upcoming support for calendar subscriptions in Microsoft Outlook 2007. With this new version every computer user will be able to use one of the desktop calendar products like

• Microsoft Outlook 2007
• Apple iCal
• Novell Evolution
• Mozilla Sunbird

or Google Calendar. There are other web-based calendar applications as well, but they will have a hard time to compete in terms of usability no matter how much AJAX-juice they use.

What is the consequence for creators of date-sensitive applications? The consequence is that you do not have to build a calendar interface on your own. All you need is to provide a date entry field with an appropriate widget and a subscription link for the calendar. Your users will be able to view all events, those resulting from your application and all other events that are of importance to the user in their favorite calendar application. This will be a tailor-made calendar application that your own calendar interface will hardly match in terms of usability, desktop integration and familiarity.

Gaurav Oberoi of Billmonk just commented on the entry How to manage book lending and borrowing web-2.0-style? that Billmonk now features public library sharing. More details can be found at the billmonk weblog.

Of all three applications (Billmonk, LibraryThing and Stuffopolis), Billmonk is the most polished one and the one I'd like to use for borrowing and lending books, but when it comes to social application features like tagging and finding taste neighbors, LibraryThing is clearly superior.

There is one important feature I am still missing from the Billmonk website: It is an RSS feed of recent changes made to my library. Together with my friend's libraries this can be aggregated into a common library feed for a team.

In response to my yesterdays post How to manage book lending and borrowing web-2.0-style? Nicolas Delsaux recommends LibraryThing. The beta geek in me registered at this service and from what I have seen, it looks good. The web design is not 1997-style, you have proper tagging, it is easy to find readers who have very similar reading lists and the other user who has exactly the same book as you. It supports many data sources, among them the amazons, the library of congress and other libraries.

Another nice feature is the bookshelf view to your book list, which is almost what I had in mind when I wished an Javascript-resizable Bookshelf view. Unfortunately there is no lending/borrowing feature in this application, which makes it hard to use for my purposes.

"The Library"
Originally uploaded by The Department.

I've been looking for an web application to manage my book collection and to keep track of lending and borrowing. So far I have been looking at:

• Ning's Bookshelf, which looks nice, but has no lending support


• Billmonk, which has no support for making your library world-visible and is limited to 99 items.


• Stuffopolis, which does not look at nice, but it has a public page for my books, offers tracking of my books via google maps and supports delicious library imports

But I still see some space for improvements for this service:

1. Revamp the tagging interface to look more like del.ico.us


2. Have AJAX autocompletion support when adding items. The first datasource to query would be the database of books already added by other members


3. Integrate with Rapleaf to keep track of a borrower's reputation.


4. Create an AJAX-resizeable view of the bookshelf like Delicious Library offers


5. Offer last.fm-like recommendations for interesting books based on shelf similarity


6. Integrate with online shops: If I am looking for a book, I will first query my Stuffopolis neighborhood if this book is available for borrowing, if not there will be links to amazon.(com|de)


7. Provide a Greasemonkey script that will do the reverse for amazon.(com|de). If I am about to buy a book there, notify me if this book is available in my neighborhood.


8. Allow me to find possible lenders or borrowers based on location, similar interest, and degree of conectivity (like openBC or linkedIn)

Update: Another improvement would be the integration of Google Calenadar to remind me and my friends of the return date for books and other borrowed items.
Another Update: I've added a follow up: Managing Book Lists - Part II and Billmonk with public library.

That Instant
Originally uploaded by ocellnuri.

I've come accross a very cool new Web 2.0 application that allows management of micro-debt and micro-credit. cyclr.com allows you to track for example payments made for your colleagues when eating outside and only one colleage decided to pay.

Cyclr will analyze the credit-debt-relationships between all users and identify debt-cycles that can be resolved without any payment. For example if Peter owes Paul 5 $ and Paul owes Mary 5$ and Mary owes Peter 5$ no payment is neccessary.

Today I read about a new, excitying service that allows you to upload your OPML files (list of RSS subscriptions or reading lists) and compare your subscriptions to similar subscription lists by other people, allows to find out what are the feeds most people are subscribed to and who are the people that subscribe to the most feeds.

The service is Share Your OPML by Dave Winer, and it is exciting, but not new. It is just a relaunch of an at least two-years old, but nevertheless cool service. It will be interesting to see how this service evolves as OPML becomes increasingly popular and more feed readers like Blogbridge are starting to support it in an intelligent way.

• Top 100 Feeds
• Most prolific subscribers (I'm currently rank no. 70)

What I am really missing is a recommendation feature like last.fm's recommendations that are generated from my musical neighborhood by analyzing the x people with the most similar subscription lists and recommending all feeds that I am not subscribed to.