No public Twitter messages.

Facebook
RSS

Another great leap for open source

Jan - 27 - 2008
Nicole C. Engard

I don’t where to begin with this. I have just read several different blogs/emails/releases all about this amazing leap for the open source ILS & other open source library tools.

Let’s start with Roy:

So anyway, here’s the skinny: IndexData, WebFeat, and CARE Affiliates have partnered to create a service they’ve dubbed OpenTranslators. In a nutshell, this service makes any of the some 10,000 databases for which WebFeat has developed search connections available to be searched through SRU or Z39.50. In [one] stroke, they have made all of these sources available for searching by any application that can work with one of these protocols.

Next, Sebastian:

I don’t mind saying that this solution also addresses one of my deep, long-term beefs with the present metasearch market. Traditionally, database gateways have been hidden behind closely held proprietary APIs, and generally each vendor has painstakingly developed their own set of gateways. This practice has set the bar of entry into the market very high and has effectively stifled competition. It has also meant that resources that could have otherwise been expended on innovation and better user experiences have instead been wasted on redundant database connector development. Our interest is in developing new, exciting applications, and supporting others who are coming up with cool stuff and new services.

And lastly, the press release:

OpenTranslators will allow libraries to use the federated search interface of their choice to access over 10,000 databases using SRU/SRW/Z39.50. The databases consist of: licensed databases, free databases, catalogs, Z39.50, Telnet and proprietary databases. Libraries that already have a Z39.50 client in their OPAC will be able to connect to, not only library catalogs, but also thousands of additional databases. Those libraries that are building or already using an open source federated search tool will now be able to expand the world of information that can be accessed. Finally, for those institutions/organizations building new mashup clients, this will allow them to access and use vast amounts of additional content.

This is such a big step for libraries and open source! At least – I think it is ;) This is certainly something I will be keeping an eye on.

[update] It seems that I (and many others) may have jumped the gun on excitement. Check out some of the valid questions made by Sol at the Federated Search blog. I will be doing a lot of reading on this in the following week and see if I can come to understand it all just a bit better. [/update]

10 Responses so far.

  1. Gerard says:

    That’s what I said too, at first. Then I read the whole thing again, and again. And my conclusion, even though I still think this is a big step forward, is that it is not by far open source related.

    The name OpenTranslators does give you that idea, but it is not open as in free. It is a service, which you can subscribe to, if you got the money for it.

    Still a good development though, as a WebFeat customer I know how good it is to have access to all those translators.

  2. Gerard says:

    That’s what I said too, at first. Then I read the whole thing again, and again. And my conclusion, even though I still think this is a big step forward, is that it is not by far open source related.

    The name OpenTranslators does give you that idea, but it is not open as in free. It is a service, which you can subscribe to, if you got the money for it.

    Still a good development though, as a WebFeat customer I know how good it is to have access to all those translators.

  3. Nicole says:

    I think I was thinking more along the lines of being good for open source products – not open source in and of itself. By making the connectors open it makes it easier to build into open source apps – but I too will re-read and make sure I understood correctly.

  4. Nicole says:

    I think I was thinking more along the lines of being good for open source products – not open source in and of itself. By making the connectors open it makes it easier to build into open source apps – but I too will re-read and make sure I understood correctly.

  5. Carl Grant says:

    CARE’s definition of open is more along the lines of what Nicole is saying. We felt by adding an open standards layer (from Index Data) on top of the proprietary translators (from WebFeat), we would give libraries more open choices in the metasearch tools they used. They could select from any open source tool, or even a proprietary tool, if they wanted, but the choice would be open to them. We’re trying to open up the technology wherever possible, whenever possible and certainly we use open standards and we try to use open source software where it exists or makes business sense. But it isn’t always possible, for instance translators are a machine and people intensive task — nothing yet exists or has been produced that would do the equivalent task in the OSS environment. So, sometimes you have to marry things (Open Standards, OSS and proprietary) in order to produce a working solution for the end users and libraries. That’s what we did. Is it open? Clearly, we thought so, but I guess it depends on how strict your definition is. Ours is defined by what we perceive to be customer needs.

  6. Carl Grant says:

    CARE’s definition of open is more along the lines of what Nicole is saying. We felt by adding an open standards layer (from Index Data) on top of the proprietary translators (from WebFeat), we would give libraries more open choices in the metasearch tools they used. They could select from any open source tool, or even a proprietary tool, if they wanted, but the choice would be open to them. We’re trying to open up the technology wherever possible, whenever possible and certainly we use open standards and we try to use open source software where it exists or makes business sense. But it isn’t always possible, for instance translators are a machine and people intensive task — nothing yet exists or has been produced that would do the equivalent task in the OSS environment. So, sometimes you have to marry things (Open Standards, OSS and proprietary) in order to produce a working solution for the end users and libraries. That’s what we did. Is it open? Clearly, we thought so, but I guess it depends on how strict your definition is. Ours is defined by what we perceive to be customer needs.

  7. Carl your definition of open isn’t open at all. It a standard layer yes, but in no way open. By your definition of open anyone offering a web service layer based on SOAP or REST would qualify as an open service. Furthermore CARE is being careless with terms hoping to confuse the market about what open is. Your service is in NO way open at all. BTW there are gobs of tools for building translators in the open source world – In fact the open source software development process is very “people intensive.” Making the connector open under the Free Software Foundation definition is open.

  8. Carl your definition of open isn’t open at all. It a standard layer yes, but in no way open. By your definition of open anyone offering a web service layer based on SOAP or REST would qualify as an open service. Furthermore CARE is being careless with terms hoping to confuse the market about what open is. Your service is in NO way open at all. BTW there are gobs of tools for building translators in the open source world – In fact the open source software development process is very “people intensive.” Making the connector open under the Free Software Foundation definition is open.

  9. Actually, this isn’t Carl’s definition of ‘open’. Open Standards and Open Systems predate the Open Source Software ‘movement’, and these interpretations have coexisted peacefully for years! I don’t see a conflict at all, and I don’t believe that Carl has ever suggested that the OpenTranslators were themselves open source — although the tools upon which the service is built are indeed OSS — you can download them from my website. :-) OpenTranslators earns the right to the ‘open’ nomer because it supports not just any old SOAP/REST protocol but a range of application-level protocols maintained or supported by NISO and ISO, and widely implemented in the library community.

    Carl and I have both characterized the service as a powerful enabler for OSS projects. Just consider LibraryFind (http://libraryfind.org/), or any of our own OSS tools and gadgets, which can be used in conjunction with OpenTranslators to access a significantly larger set of resources. Today, only a very small number of content providers support standards-based interfaces. Would Open Source translators be better if we had them? Sure, but not nearly as good as Open Standards interfaces to all of the resources we’d like to search! We actively promote and support both, but in the meantime, this product allows us all to access resources that would otherwise have been unreachable without considerable effort.

    PS: For a definition of Open Standards, refer to http://en.wikipedia.org/wiki/Open_standard . If you search in the text, you will find several references to Open Source Software and the synergetic relationship between the two (I swear I didn’t edit the WikiPedia article to support my argument!! :-)

  10. Actually, this isn’t Carl’s definition of ‘open’. Open Standards and Open Systems predate the Open Source Software ‘movement’, and these interpretations have coexisted peacefully for years! I don’t see a conflict at all, and I don’t believe that Carl has ever suggested that the OpenTranslators were themselves open source — although the tools upon which the service is built are indeed OSS — you can download them from my website. :-) OpenTranslators earns the right to the ‘open’ nomer because it supports not just any old SOAP/REST protocol but a range of application-level protocols maintained or supported by NISO and ISO, and widely implemented in the library community.

    Carl and I have both characterized the service as a powerful enabler for OSS projects. Just consider LibraryFind (http://libraryfind.org/), or any of our own OSS tools and gadgets, which can be used in conjunction with OpenTranslators to access a significantly larger set of resources. Today, only a very small number of content providers support standards-based interfaces. Would Open Source translators be better if we had them? Sure, but not nearly as good as Open Standards interfaces to all of the resources we’d like to search! We actively promote and support both, but in the meantime, this product allows us all to access resources that would otherwise have been unreachable without considerable effort.

    PS: For a definition of Open Standards, refer to http://en.wikipedia.org/wiki/Open_standard . If you search in the text, you will find several references to Open Source Software and the synergetic relationship between the two (I swear I didn’t edit the WikiPedia article to support my argument!! :-)


Bookmarks for July

Today I found the following resources and bookmarked them on ...

Bookmarks for July

Today I found the following resources and bookmarked them on ...

Bookmarks for July

Today I found the following resources and bookmarked them on ...

Bookmarks for July

Today I found the following resources and bookmarked them on ...

Bookmarks for July

Today I found the following resources and bookmarked them on ...