http://www.sqlservercentral.com/blogs/the_future_computing_appliances_and_sql_server/2010/06/26/the-_1820_addicted-to-data_1920_-decade-is-still-in-its-infancy/ Printed 2014/04/24 06:11AM
The ‘addicted to data’ decade is still in its infancy
“Content is king” is the golden rule of the SEO. The semantic web is all about data. Any web or traditional application has no value without data. Few people have discussed the future of Data, Pat Halland being one of the most prolific ones. Data is and will be everything:
- An internet scale document living on the cloud has significant differences to the design for 140 characters. In addition to the vast amounts of information being produced, it’s getting harder to get “all the data” stored on memory. This means there is never a consistent global view of the whole structure. Implementing an “eventually consistent” version of data is a hard problem that impact the design and responsiveness of future applications. For example, Microsoft Outlook implements an <offline> mode – simulating expected results on the client requires additional logic. It works very well. Most twitter clients wait for user actions to take place, instead of doing the work on the background – however the operation might actually never take place on the cloud. More on this later.
- Social computing is allowing many users to participate on collaborative creation, this means that data structures need to use more unique identifiers so versioning and consolidation can take place. There is a significant performance and UI impact for building these multi-screen applications. Envision how memory requirements would affect excel for adding multiple versions of each cell.
- The Social generation will continue releasing additional personal information, combined with location and activity. See data life of the future. A key way to combat the information overflow is to display information that is relevant on very specific moments. We should expect this functionality on future e-mail and communication clients. Acquiring, tagging and classification will continue relevant.
- It’s expected that applications will leverage information from many sources as well as information on public cloud databases (Dallas, AWS). Pay attention to the emerging OData proposal. Toby Segaran published a great presentation on Social Data analysis.
The problem goes beyond data to visualization, we can argue that after 25 year of the existence of Microsoft Word, most users cannot produce a good looking document. Three months ago I blogged on storage limits. Emerging topics also include proposals allowing cloud developers to control their data travels across devices. Yes, the Exabyte age is just arriving…