Have been doing some more thinking on the Digital Dashboard and how it needs to be put together. The thinking has also been motivated on how best we can integrate the work being done internally by three different teams as part of the BlogStreet, Digital Dashboard and Enterprise Software projects.
What the user sees is the Digital Dashboard. For now, it is just a page in a browser with three tabs (Mozilla allows tabbed browsing, instead of opening three separate copies of the same application). The three tabs are for:
– the RSS Aggregator
– the home page of my Blog (Reading)
– the New Entry window (Writing)
The RSS Aggregator needs to have the following features:
– add/edit/delete RSS feeds
– show the items (viewing): this can be done as an Outline
– support for multiple pages, so I can segragate feeds
– automatically set up filters to route specific feeds
– stored in a database
– filter by source, time
– show only new items
– search in the archives of the RSS feeds (which means storing old items)
The interesting feature here would be the “auto-route” filters. One way to look at RSS feeds is to look at what we want to do with email. Filters are an important part of mail management. In the RSS case, the auto-route filters should be able to do more than just posting to a blog or re-sending to email/SMS. We need a simple language to take certain actions on the items coming in. Sieve used in the Cyrus IMAP Mail Server could be a good starting point. But we may need more rules than what Sieve provides.
Today, people can subscribe directly to an RSS feed. This may need to change, especially in an organisation. I should be able to restrict who can subscribe to my feed. This becomes somewhat like IM – both sides have to accept to communicate. This will enable me to distribute feeds selectively. A person’s email address may work as authentication (just like many listservs do for subscription to email newsletters).
It may also be possible to build an “RSS market / exchange” where RSS feeds are aggregated together from different sources. A market approach also gives the option of charging for subscriptions. It also enables creation of feeds and their availablity for sources which don’t have the feeds available at this point of time.
An interesting feature in RSS would be to “overwrite” an existing item if unread. For example, a sports website could periodically send updates of the score. It the last item hasn’t been read by the user, then the new item should “in-place” update the old item, so that the user does not end up with a lot of items. This is somewhat similar to the ideas that Zaplet had when it had launched its product (it did the updates in place in email).
The next key component is the Blog. I should be able to do the following:
– post to multiple blogs
– each blog should have multiple categories
– blogs can be individual or comunity managed
– posting should require some authentication
– each blog should publish its own RSS feed, with the entries and the comments
– in an enterprise, do blogroll and link analysis to automatically identify clusters
– upload the blog to another location for hosting
This is the underlying infrastructure for putting in place the Digital Dashboard. The one platform which comes close with many of these features is Userland’s Radio which costs USD 40 a copy. In the open source world, there are many components, but what’s not happened so far is the seamless integration of the entire chain.
Once this infrastructure is in place and users begin with feeds coming in from various news sites and bloggers (within and outside the organisation), it will be time to “switch on” the next upgrade: the ability to route events from business applications through the same RSS-Blogs network. This combination becomes the foundation for the real-time enterprise.
The Digital Dashboard ideas are akin to today’s Corporate Portals. The problem with the latter is that they are expensive and mainly focused on the bigger organisations. What’s needed is a simpler, grassroots platform the the mass market of enterprises.
This is also the way for the Linux Desktop to make inroads into the enterprises. By just asking people to switch applications (Windows to Linux, Microsoft Office to StarOffice or OpenOffice), there’s going to a huge reluctance because people are used to the applications, the keyboard short-cuts, etc. What’s needed is a change in the way people interact with information and computers – which really is what they are supposed to be doing. It is the way Nvidia is switching the game from CPUs to GPUs. The Digital Dashboard needs to become the new Enterprise Operating System, the new Desktop for people.