OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

semantic-ex message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Experiences in discovering Web services (action item from lastcon-call)


Dear SEE members,

 

One of the action items from the last SEE phone-con was for me to email the list with the experiences I had in searching for a useful set of Web services on the Web. Although the searching was informal, it became increasingly obvious that there are very few public WSDL web services that are not demonstration examples or that are only accessible once you have some kind of sign-in credentials. Theses credentials were usually only available by completing an online registration form. A report with empirical data on using different approaches to service discovery is available at [1]

 

For my investigation, I wanted to see if there was a useful domain (in the context of a B2B integration paper I was writing with some others) that had a number of realistic invocable Web services, without the need for credentials. I tried two approaches:

 

1. I examined a set of WSDL URIs harvested by Holger Lausen and Juergen Umbrich from DERI Innsbruck. From the URI of the WSDL, many could be discounted straight away as test or dummy services. I then broadly categorized the services and then used a tool called StrikeIron to see if I could invoke the service and get some kind of response.

 

2. I used the Google filetype search mechanism to search for WSDL file types along with keywords that corresponded to my rough categorization from the first step. Any interesting WSDL I found, I again tried to invoke using the StrikeIron tool.

 

 

The conclusion:

 

Well, this was an informal experiment but it seems clear that there are relatively ‘real’ WSDL Web services that are publicly available on the Web. I did not look into any UDDI registry. XMethods [2] provides the biggest browsable list of services. that I came across. The most popular domain areas for services were (in no particular order) weather, GIS/geographic/ mapping services, genetics, parcel-tracking (usually need a sign-in), currency conversion and SMS messaging.

Going through the process made me question the motivation for heavy investment in semantic service discovery. I see the benefits mainly in closed environments, especially in larger organizations where the services are published inside the firewall and are only available to SOAs inside the company’s boundary.

 

 

[1] Daniel Bachlechner, Katharina Siorpaes, Dieter Fensel, and Ioan Toma. Web Service Discovery - A Reality Check. Technical report, Digital Enterprise Research Insitute (DERI), January 2006. http://www.deri.at/fileadmin/documents/DERI-TR-2006-01-17.pdf

 

[2] http://www.xmethods.net/

--------------------------------------------------------

Matthew Moran

Digital Enterprise Research Institute

Phone:     +353 91 49 5017

Mobile:    +353 86 196 5648

Fax:         +353 91 49 5541

DERI Web: http://www.deri.ie

Homepage: http://sws.deri.ie/members/matt

--------------------------------------------------------

 



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]