Enterprises, Struggling to Manage Your Data? Try The Semantic Web

Written by Richard MacManus / May 29, 2009 5:00 AM / 10 Comments

A new PricewaterhouseCoopers Technology report explains how the Semantic Web and Linked Data can help enterprises manage their large-scale data better. The PwC Center for Technology and Innovation team spent several months researching and analyzing the problem of data silos in enterprises - and what solutions are being developed to help with that problem. The answer, according to PwC, is Semantic Web techniques. PwC believes that the Semantic Web offers a practical way to address the problem of large-scale data integration.

We downloaded the 58-page report and summarize some of the findings for you in this post.

PwC wrote in the report that the underlying technology of the Semantic Web applies not just to online data, but to "internal information and non-Web-based external information" - including a company's data warehouse. Currently enterprises struggle with hefty relational databases, but PwC says that the Semantic Web could both lower costs and provide more data. What's more, companies can contribute their "non-sensitive" ontologies to the Linked data cloud:

"With the Semantic Web, you don't have to reinvent the wheel with your own ontology, because others, such as musicontology.com and DBpedia, have already created ontologies and made them available on the Web. As long as they're public and useful, you can use those. Where your context differs from theirs, you make yours specific, but where there's commonality, you use what they have created and leave it in place. Ideally, you make public the non-sensitive elements of your business-specific ontology that are consistent with your business model, so others can make use of them."

We recommend you read the report for a full understanding of ontologies, but basically they are the means to connect data and metadata together. PwC goes on to explain that ontology development is becoming a more popular business-integration technique for organizations. It cites the data-intensive healthcare, media, and oil and gas industries as being early leaders in this movement.

The problem that enterprises often face with Web technologies is the lack of control they introduce. PwC acknowledges this, but says that enterprises only need to control some data - not all of it. To take true advantage of the Web, says PwC, enterprises need to be able to take advantage of its scale - and that's not possible without giving up some control. So PwC advises enterprises to "limit your controlled environment to what you don't have to scale."

The table below is a great summary of the benefits of the Linked Data approach to data management.

PwC lists a number of Semantic technology vendors that focus on enterprise data integration techniques. Not all of these solutions are based on Semantic Web standards, but most have at least a partial adherence to them. Some of the vendors that PwC lists include Cambridge Semantics, Collibra, Metatomix, OpenLink Software, and Talis Group. Interesting to note that some of the big database vendors are also in play in the nascent Enterprise Semantic Web market: Microsoft has an open-source Ontology Add-In for Office Word 2007 (among other products); and Oracle offers Spatial 11g, an option to Database 11g that according to PwC offers a wide range of semantic data management features.

The report, available as a free download at PricewaterhouseCoopersTechnology Forecast, is packed with useful information for enterprises wanting to make a foray into the world of the Semantic Web. Tell us in the comments if your company is using the Semantic Web, and how it's going for you so far.

Comments

Subscribe to comments for this post OR Subscribe to comments for all ReadWriteEnterprise posts

Hi Richard,

thanks for sharing this report. I'm surprised and euphoric at the same time that PwC seems to have nailed it to the point. Linked Data in enterprises is a fantastic opportunity (which I've been exploring for a while now).

And of course it is flattering to see my DBpedia mentioned :)

Cheers,
Georgi

Posted by: Georgi Kobilarov | May 29, 2009 6:07 AM

Here's flexible way to specify and run applications over linked data.

There is a system on the Web that works as a kind of Wiki, for content in open vocabulary, executable English.

English text (like this sentence) is normally something for a person to read, but it cannot be used as a program that you can run on a computer.

On the other hand, executable English is something that a person can read, and that you can also run on a computer.

As befits a Wiki, shared use of the system is free. Just point a browser to www.reengineeringllc.com .

Since the executable knowledge is in English, Google indexes and retrieves it, acting as a kind of registry.

You can use your browser to write programs in English, run them, and get English explanations of the results.

Posted by: Adrian Walker | May 29, 2009 7:00 AM

Great post! Thanks.

Posted by: François Dongier | May 29, 2009 7:13 AM

Excellent report by PWC and synopsis by RWW.
I think the next frontier will be in light-weighting traditional ontology-driven approaches and deriding them from the rigidity and latency they inherently come with. (using Linked Data is one step in that direction, and it would be good to see real use cases showcasing the full-cycle benefits).
Also, adding bottom-up user-driven approaches for enriching and evolving taxonomies will be an interesting development to watch, and one that will accelerate the adoption of Semantic web principles into the mainstream.
As for the enterprise-side, contexualizing external linked data and re-integrating it into existing corporate content systems is not a trivial task.

Posted by: William Mougayar | May 29, 2009 7:17 AM

very well, thanks...

Posted by: Hasan Aydın | May 29, 2009 12:49 PM

This is a great article. The semantic web aka web 3.0 aka the next generation is truly where the internet is heading. It's amazing when you stop and think about where the web has evolved to in the last 10 years. Kids going into college now have such a competitive edge with technology. It's truly amazing the capabilities of computers and our wonderful internet world!

Posted by: Brent Grablachoff | May 30, 2009 10:48 PM

That PWC paper's primary argument is that relational database and SQL are apparently too rigid, siloed, and require too much effort to maintain and create either reports or tools and process for users to have flexibility and ease of use.

Well, the root cause for these shortcomings has always been primarily due to lack of engagement, leadership and ownership by business in the standardization of definitions of business objects (people, services and interactions) and incorporation into the database and application.

The semantic web won't just happen because it is implemented. Someone will have to wrap data with semantic definition. But this has to be done for both relational database/SQL or the "semantic web". I haven't seen anything in this PWC paper that addresses this essential requirement nor the practical considerations of how this is done.

Of course, universal definitions for data is obviously of immense benefit. However, there is still a lot of value to be wrung from relational databases, SQL and current application technology. But it needs the business in a leadership role in ensuring that data definitions are incorporated in either case.

The only way to do this is to have people who build and maintain databases and applications who understand the business, its strategic goals, and how the business process needs to be translated into and supported by relational databases, applications and reporting.

Nothing new there!

Posted by: canadianalien | May 31, 2009 1:12 AM

@canadianalien

You _can_ semanticize relational/tabular data, but then it wouldn't
work so well when everyone else is semanticizing RDF/graph data.
It is not as straightfoward as some people think to simply map
relational data onto a graph/RDF model.

This is the reason we have standards such as RDF--so that people
who follow them can interchange data. By sticking with the 30 year
old relational model, you sacrifice the very thing standards are
for. The only benefit of traditional databases is the fast querying.

I understand though that some people who have worked in this field
their whole life resist change.

Posted by: Al | June 3, 2009 10:12 AM

This is the reason we have standards such as RDF--so that people
who follow them can interchange data. By sticking with the 30 year
old relational model, you sacrifice the very thing standards are
for. The only benefit of traditional databases is the fast querying.

Posted by: cheap air jordan retro shoes | August 3, 2009 5:37 AM