MLS agnostic… unless
I have made my rounds at the conferences advocating MLS agnostic technology. I still do. I firmly believe that the future will see more and more tools, widgets and technologies developed that interface with MLS(s) but are not dependent on or tied to a particular MLS… they are MLS Agnostic. The advantages for the Association technologist are obvious, as they are for the practitioner. Content can be syndicated between MLS applications, taking the ‘mashup’ to a whole new level.
Imagine a listing input widget that could push your listing input to virtually any MLS you point it at. Sure, not all the fields and data are standardized, so it does the 90% of the work that it can and then you spend an extra 60 seconds backfilling the deltas for each MLS you push to. In the end, you have saved yourself time and effort, AND you have the original listing input object in your possession. If you want YOUR listing content back, you don’t have to ‘borrow’ it back from the MLS.
I believe in this so much that I put my development efforts where my mouth is, and will continue to do so. In a world of shifting vendors, regionalizations, etc. it just makes good sense not to invest too deeply in a technology which cannot be retooled for future interoperability with an MLS at worst, and is not natively MLS agnostic at best. I will try to continue to follow this path, unless…
Open Source MLS where art thou?
Open source, if you don’t know, is cooler than the other side of the pillow. Open Source is the software movement most commonly identified with its most successful progeny Linux. In essence developers take their work product, often built on the foundation of others, and ‘give’ it back to the community, with the only licensing requirement commonly being that the next person in line do the same. It allows millions of developers all over the world to ‘check out the code’ and work with it, extend it, build plugins or additional features and support, and then ‘check it back in.’
One of my proudest accomplishments in my role has been to develop and deploy an open source RETS ruby client. Other developers have already picked up the application and extended support to other version of the RETS standard than it was originally coded for, in essence giving value back for free and making the application more MLS agnostic (it can support more MLS’s).
This obviously leads to the question - will there be a genuine open source MLS application someday?
Most discussion of this in the industry has focused on open content access standards (free and ubiquitous access to listing content - an open database so to speak). See Michael Wurzer’s excellent blog on this point. This is NOT what I am discussing.
I am referring to the actual application itself. I am referring to the ability for millions of programmers around the world to take a crack at their idea of the perfect CMA module, or property valuation algorithm, or search function, and on.
Now that could be cool - if the MLS itself were agnostic. The advantages for practitioners and in turn consumers seem obvious. Can you think of some of the pro’s and con’s I may have overlooked? There are several I can identify offhand, but I don’t want to rob the reader of the satisfaction of identifying their own.
I do want to thank the folks over at Clareity for bringing my thoughts back to this topic. Gregg Larson, Clareity CEO and Matt Cohen, both thought leaders in the industry, gave everyone something to think about yesterday as they weighed in on the CAR statewide MLS proposal with some thoughts that may have been overlooked, such as “Facilitate statewide open and free data standards for listings and membership data,” “standardize statewide MLS rules and regulations,” and “facilitate the creation of a ‘virtual’ statewide aggregation of MLS active and comparable data that can be accomplished using existing infrastructure and technology by existing entities.”
When I see those thoughts in alignment, for some reason I see the possibility for not just MLS agnostic technologies that sit between MLS applications with more free and open data standards, but the possibility for an Open Source MLS application built on the results of these sort of advances. Others may see something very different (and I should point out that Clareity themselves are not advocating this)… but that is the beauty of the open standard - everyone is allowed to realize their own vision
Michael - Good article. Where can we get our hands on the open source RETS Ruby client?
The jury is out on whether someone should open source or not - different strokes for different folks - but let the competition commence!
Dan, thanks. The project is open at rubyforge here:
http://rubyforge.org/projects/rets4r/
Many compliments to Scott Patterson the primary author.
Best,
Michael
October 16, 2007
I’ve added the link to the RETS wiki at http://wiki.rets.org/wiki/Opensource#Ruby Please feel free to enhance the description of the tool. Thanks for making this available.
Great article, Michael. I think listing maintenance is the best case for open source, with focus on data validation routines such as geo-coding, address validation, tax records confirmation, etc. This leans into my suggestion that RETS focus on trying to define a universal identifier for parcels/properties and also into the thoughts you all have expressed on this blog about providing brokers control over their listings. If listing maintenance is in their control, then the only question is how aggregation occurs and that’s more of a business question than a technical one.
December 15, 2007
very interesting, but I don’t agree with you
Idetrorce
Comments RSS TrackBack Identifier URI
Leave a comment











6 Comments