[ontology-summit] [Making the Case] To Small Businesses and Startups

[Bart Gajderowicz <bgajdero@xxxxxxxxxx>]

This is a side note to the one discussed at today's telecom and Michael Uschold's "Blank Stares" thread:

http://ontolog.cim3.net/forum/ontology-summit/2011-03/msg00088.html

I want to focus on technologies as a ready-to-use example of semantics.

Much of the talk and the examples so far have been centered around enterprise use cases or large scale systems. I deal with startups a lot (big tech start-up community in Toronto) and their priorities are of course very different from those of larger companies. These companies often look for a solution that gives rewards quickly, and that usually means quick, dirty or both. Often what is needed is a proven track record and a set of tools that make it quick to implement a new technology, and have it integrate with their existing system. The moment I mention semantics I get responses criticizing RDF, triple-stores, lack of adoption, the point of linked-data, etc. These are all implementation details, and have little to do with actual semantics or ontologies. The Semantic Web has put a cloud of confusion around semantics because so many technologies and solutions are grouped under that banner. As helpful as SW has been, it has also been a draw back too.

What's the problem with small companies adopting semantic technologies? These companies:

- won't necessarily have the resources to take on the amount of work needed to analyze a new technology and integrate it into their current system. 

- don't need high level of detail, only what is needed to develop the features that will get them the next round of funding.

Why is this important?

- an Open Source solution is always the preferred option. 

- the success of Open Source technology depends on their adoption by the developer community. 

- the more developers involved the better.

This also applies to Open Data. Although many datasets come in the form of RDF, the majority are still either a relational DB, a denormalized table, or a spreadsheet. The number of free or pay-per-use OpenData repositories is growing, and their adoption of semantic tools is important as well.

While there are many open source projects, the response I get is that many of them are still "too academic". For example I use Protege and the OWL API almost on a daily basis, but it's overwhelming for even technical people. Despite of a small company's agile environment, the main focus is always on

- current key technologies

- developing features

- get funding

- grow user-base

It is often much easier to simply code something up in their current environment to "mimic" semantics and use their existing DBMS.

Responses to Arun's question on the following thread "An example of the worth of ontology development" gave some examples of technologies to convert data to ontologies:

http://ontolog.cim3.net/forum/ontology-summit/2011-03/msg00035.html

Toby Segaran's slides provide a good selling point on how easily semantics can be applied to data via triples.

http://kiwitobes.com/presentations/SegaranCSHALS2011.pdf

We need different levels of technologies for open source adoption. Small companies and startups are dying to get on any new technology's band wagon, but only if they can use it quickly and cheaply with maximum results. 

--
Bart Gajderowicz, MSc.
Ryerson University
http://www.scs.ryerson.ca/~bgajdero

_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/   
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/  
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2011/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2011  
Community Portal: http://ontolog.cim3.net/wiki/     (01)