Last week I asked my colleagues on the W3C staff to predict Web trends in 2009. I compiled (with some editing) a few of the predictions I received. Naturally, these informal utterances are not endorsements and do not represent W3C as a whole. There are also important topics not covered here (such as the impact on accessibility of yesterday’s publication of WCAG 2.0).
Let us know what you think is likely to happen on the Web in 2009.
Blogs about data visualization are popping up everywhere, and we are seeing more software with easy to understand syntax. As one indicator, the New York Times has created a dedicated team of designers for visualization, see their blog.
Linked [Open] Data
The linked data movement will continue to grow, and we will see a growing number of real-world linked data applications (beyond the techie circles). The seed of a genuine Semantic Web of data has been planted and will germinate in 2009. The recent ‘join’ of freebase to linked open data (LOD) is one sign.
A typical application is to help in combining the messy, scruffy, but (deceptively) simple application of tagging with a slightly more ordered (and therefore more scalable) usage of generic URIs. For example, social bookmarking will automatically analyze the content of your entry and will propose reasonable tags; these tags may in fact be LOD URIs, i.e., they will refer to some real data that applications can much better exploit. Linked Open Data is here to stay.
Furthermore, the number of triples that can be stored and managed will continue to grow by leaps and bounds. This is important because large application domains (HCLS, Oil and Gas/Energy, Financial Services, etc.) rely on data integration with huge amounts of data. This raises the possibility that new and important communities will be able to use this technology. The effect of this scaling may be visible by the next Semantic Technologies Conference. Note that some of the new OWL work (i.e., some profiles) is essential for these large scale triple storage environments, and the same is true for rules. Some upcoming W3C standards in this area will coincide with this evolution in linked data. And some areas of suggested new work (SPARQL+, relationships to relational databases) relates to this as well.
What’s missing? Privacy and security. It is extremely important to develop those areas. For instance, W3C is involved in the PrimeLife European Project, which touches on privacy issues. Keep an eye out for distributed reputation systems.
Open Government Data
We may see “Open Government Data” become the de facto standard in demand by an increasing number of civil societies, a topic discussed within the eGovernment Activity at W3C. See, for instance, in a draft set of open government data principles, which includes: “Open [government] data promotes increased civil discourse, improved public welfare, and a more efficient use of public resources.”
As a recent example, “Open Government Data” ranks high among recommendations at the community-driven Barack Obama CTO site, which compiles ideas for the upcoming United States Federal Chief Technology Officer.
Can I Use That Content? – Policy Language Interoperability
Creative Commons has been a big success helping people determine what they can do with content. Creative Commons recently brought CCrel to W3C, a technology intended to make copyright statements machine readable, searchable, selectable. The goal is that people will have an easier time helping others reuse and circulate content. W3C has some experience in this area (PICS, P3P).
Beyond the realm of copyright, we may see the impact of machine-readable statements affect the distribution of commercial fonts, which have not yet flourished on the Web. One traditional obstacle has been that users do not have easy access to font license information — you typically cannot “right-click” on a font to see who holds the copyright. It will be interesting to see if the market for fonts opens up as metadata and supporting software increasingly help users make more informed decisions.
W3C’s Policy Languages Interest Group is looking at issues of privacy, access control, data handling and other important aspects of data governance. PLING was created in 2008 in response to a W3C Workshop where participants observed fragmentation in the market for policy languages. The PrimeLife Project (mentioned above) has begun to contribute interesting insights to this forum for discussion of policy languages and how to make them interoperable. Talking about privacy without talking about identity may prove difficult to handle, however.
Mobile Web and social networking will continue to drive the need for improved tools and good practices for managing privacy and safety online, especially for young people.
A number of trends suggest that critical mass for location-based services (such as search) may be at hand:
- GPS receivers are becoming quite affordable and they even come built-in to other devices, such as mobile phones.
- EXIF and XMP both support longitude/latitude now and there is a choice of software that does interesting things with photos that are thus tagged with their location.
- Vcard address data can contain longitude/latitude information (which is thus also available in HTML via the hCard microformat).
- The geo microformat is under development to allow HTML pages as a whole to be linked to a particular location on earth.
- Google Earth is becoming useful as a tool to search for location-dependent information (photos, wikipedia articles, etc.); soon it will have so much information that it will need a new interface/concept to filter the less useful data.
- Sites such as eventful.com allow searches based on distance.
As mentioned above, it will be interesting to observe what mechanisms emerge to help manage privacy and security issues in this space. On the W3C end, the Geolocation Working Group was chartered to “define a secure and privacy-sensitive interface for using client-side location information in location-aware Web applications.”