A standard, like software, is never complete - it just goes on to the next version. But, the RETS standard has been going on for ten years - I think it's time to take stock about how far it has come and talk about where it could or should go.
If you're an MLS executive or an agent, why should YOU care about this RETS stuff moving forward quickly? Doesn't it already facilitate your data feeds, syndications and MLS collaborations? Think about this - recently I was negotiating an MLS contract and wanted to know what data, created by my client and their subscribers could be moved from the old system to the new one:
Listings? Of course.
Roster? Of course.
Contacts? That might be possible with custom work, not using RETS.
Prospects? No.
Saved searches? No.
Custom columnar reports? No.
Custom search screens? No.
More than half of the items that we wanted to move from system to system using RETS couldn't be done. Each one we couldn't move would mean pain (time/effort) for the MLS subscribers as they had to re-create the data in the new system manually. This isn't just relevant to moving from one system to another - it's relevant to offering a parallel system - or multiple MLS front-ends - or any number of potential MLS add-on products. And I could make similar points about the difficulty of moving between other real estate systems, such as association management and transaction management systems.
This difficulty in moving systems or adding systems (some would call this the "friction of the transaction") does not encourage the most competitive marketplace for real estate technology - and this is bad for my MLS and Association executive clients and bad for their subscribers and members. Face it - most of the people driving the data standards are more established vendors - and they do not have incentive to improve standards to make it easier to move to another system. So, if you want the standards completed to the point where you can move *all* of your data from system to system, increasing real estate application innovation, increasing competition in the technology marketplace and making transitions between systems easier for your members - YOU have got to work with your peers to drive the standard forward.
If you want to contact the Chair of the RESO Outreach and Education committee, email Kristen Carr (kcarr@big-llc.com).
The Fall RESO/RETS Conference is September 23rd - 25th, 2009 in Chicago, IL - more information is available here: http://www.rets.org/meeting
Matt, Very well said as usual. The other major gap in RETS is that we don't have an adopted common data definition. So companies like Wolfnet with about 300 MLS interfaces, have to support 299 different scripts. In addition, as MLS or their vendors change the data on an ongoing basis, all of the third party providers of broker/agent solutions like Wolfnet have to tweak their code. We can and must do better.
As someone who has worked for years within the RETS community slowly pushing the boulder uphill, it has been painful to effect any progress with regards to business functionality. Technical "improvements" aside, from a business perspective, RETS 1.7.2 serves the same narrow use cases as RETS 1.7, RETS 1.5 and RETS 1.01.
With some creativity, many folks (myself included) have found ways to extend RETS in several ways to provide additional functionality. But that varies from implementation to implementation and belies the concept of "standard", doesn't it?
The promise of RETS as an enabling technology could solve all sorts of business needs. In addition to the ones you mention above, RETS + NRDS could be a standard way to interface with membership systems. RETS + PRIA (public records std) could be a standard way to interface with county records. Years ago RETS workgroups discussed "profiles" for data replication, CMA and hotsheets. These discussions live on in hallway conversations and need to be brought back onto the agenda.
To Kevin's point about common data definitions, in 2006 the community had reached agreement on universally supporting approximately 80 core standard names, but that fell to the wayside.
The RETS servers in place today do serve an important business function. But that function only represents one portion of the Real Estate Transaction.
Paula, thank you for your reply. I fully agree that we may want to leverage various existing standards. The trick is going to be how much to simply refer to them versus incorporate them - different ones will have different answers. At some point, we'll need to incorporate (rather than just refer to) something like BRML (business rule markup language) so that every time MLSs move from system to system they don't have to re-communicate their undocumented business rules. When they don't succeed, it's frustrating to all parties - why not document via RETS? That's ONE MORE use case for RETS that hasn't been addressed. :-)
Matt, what resonated with me was the following statement: "most of the people driving the data standards are more established vendors - and they do not have incentive to improve standards to make it easier to move to another system". As an association manager, I found that fact a major stumbling block. And what existing then, and I suspect still does, is the fact that when asked if their MLS vendors provide RETS functionality, they proudly nod and say "Of course". What they don't seem to realize is that in many cases the functionality is skeletal at best, and outdated and virually useless at worst.
Judith - very few MLS vendors are actually compliant - even some of the well established ones are going through a lot of compliance angst right now. As the standard grows, it will be even more difficult - I do understand things from their perspective too. I agree with you for the most part, but "virtually useless" is perhaps a bit harsh? I do think that the standard needs more attention from the business side, evaluating possible use cases like the ones I've described in my post and subsequent comment and figuring out how to move the standard forward to accomodate them. This hopefully will be a call to action for MLS executives. I have more concrete steps in the planning stages ... stay tuned.
Thanks for another great post, Matt! As most people who know me know, I've been heavily involved with RETS for about 7 years. One thing that bothered me in the beginning was that the standard was being moved forward by technical people, not "users". While they did a great job, they did not focus on what was needed in real life, by real users. At the Austin meeting in...ugh...'07 maybe - I kind of flipped out and said we needed to get a Business side of RETS going.
That initiative got a ton of support! Before all we had was Marketing but we created workgroups such as Education and Outreach. Certification (which is being renamed to something else) and formed a Business Committee. We (yourself included) have tried and tried to get MLS operators, 3rd party product vendors, brokers, etc. to attend RETS meetings and tell the Standards (Technical) side what they need and want. It's so hard to get Execs to attend meetings because they believe it's too technical. The RETS sessions I've facilitated at NAR, Midyear, etc. are not attended very well but the people who do show up are amazing!!!! They ask questions, they teach us things. We need this people!
We seriously need the MLS Execs to tell us what they're missing, where they think RETS is failing them. It isn't okay to "settle" in this world. I don't believe the "established vendors" have any animosity toward RETS or moving data around (and I'm not implying you believe that - I know with absolute certainty that you don't believe that) but if they aren't hearing from their customers that the customers need RETS to do X and/or Y they just aren't going to push that into the specification. And I'm not talking about RETS Management Tools here, I'm talking about driving the standard to include the things you listed in your post.
MLSs, Brokers, Vendors - everyone in the industry - is welcome to come to the September meeting. We have a great panel set up and I'm hoping to have some "haters" standing there saying what they believe is broken with RETS. The Intro to RETS session is laid back - after a 20-30 minute presentation we basically just open the session up to questions and discussions. I always, always hear from a lot of the people in that session that they learned a lot.
I did a State of the Standard presentation for Matt and Gregg at the Clareity MLS Workshop and had a few people tell me they had been scared of RETS until after that session. Thanks Matt and GL for giving me the chance to take the fear away...even just a little. So all you Execs out there - it isn't scary!!! Your input is needed or the standard will continue to move in the direction the technical folks think is right, they're doing the best they can without your help!
I believe it would be of great value to have RESO at least make a draft layout of all of the data schemas you mention. This will provide at least a starting point for anyone looking to share this type of information who "want to do the right thing".
I have been dealing with this issue for years. One data type you didn't mention is open house event data. I am in the midst of aggregating open house data from MLS feeds (60ish so far) and find it to be the wild west. I even run into feeds where the date of the open house was left out of the data!?!?!
Paula and I collaborated on a very successful RETS 2 (I know, I said a bad word) based data transfer for transaction management data for the exact reason you mentioned, to prevent redundant data entry. We would scan the MLS listing RETS feed and retrieve any listings that went pending and auto generated a transaction record in the TM system. A no brainer in my mind, but viewed as "cutting edge" by the vendors on either end of the systems.
Best thing we can do is keep fighting the good fight and sharing our successes and challenges along the way so that the RE.net understands that there is movement in this area and their contribution is needed!
David - darn it! you're right - I completely gapped open houses. That's why I love blogs ... use the wisdom of the crowd.
Kristen / David - I think what we need to do to take this to the next level is not to ask MLS executives to engage us geeks on our terms at the meetings, but to actively engage with them as SMEs (subject matter experts) in the way that software efforts usually do. I think there's always been the expectation that somehow executives will come to the meeting and "ask for stuff", then somehow *someone* would take that conversation and drive a solution that fits the requirement into the standard. It doesn't work that way. As Paula pointed out, those things stay "hallway conversations".
To succeed, I think we would have to restructure how requirement gathering happens and create a path right from requirement gathering all the way through draft creation, RFC and standards approval. This type of rigor is what I have been hoping comes from the more formal structure of RESO. I haven't seen it ... yet.
Saved searches is the biggest issue you mention, and I think doing that would require much more focus on data standards. I've been harping on this for three or four years now but nothing happens.
I agree, to an extent, with your comments. First, I truly believe we have successfully mappy out the process for submitting change proposals, moving them up thru the appropriate Workgroup, then to the Standards Committee and to the Board. That piece of the process works. What doesn't work is RESO is still completely owned and funded by NAR so are dependent on NAR for funding to push the work through to publication. There are a lot of RCPs out there just waiting to get added to the docs. The Board (of which I am a member) has requested CRT approve the money we need. Hopefully that'll happen soon. And, hopefully the Bylaws will be approved by NAR and RESO can start working toward funding itself!!!!
Mike has been very vocal about data standards and I agree completely. Bridge is working on a project we're calling Rosetta which will be free to MLSs. What we're doing is mapping MLS fields to the different RETS Schemas. As soon as I can write some of the info on it I'll post it to the world.
I haven't attended a RETS meeting in a while, so maybe things have changed, but I find it interesting that the remaining schemas you identify are the "people" side of what an MLS might store. It's a different level of metadata that doesn't really involve the description of properties anymore. That is a huge expansion in what I would consider the scope of the RETS mission. Getting anything close to unanimous consent on what a "save search" or a "columnar report" resource would look like would be miraculous.
Considering the challenges involved in simply moving the Property data -- the stuff we can almost agree on what should be transported, all the remaining schemas proposed would have to go much deeper into permissions granted down on the individual agent level. All sorts of opt-in/opt out scenarios would then exist that an MLS would have to code for and maintain, as well as educate every single member on the implications of what their selections might be. This goes way, WAY, beyond the broker opt-in / opt-out levels that I've seen implemented.
Let it be known, I am a huge, huge fan of RETS, and live in it darn near every day, But to be honest, I don't envy anyone that has to implement those remaining schemas and the privacy responsibilities that go along with it. Bearing in mind that most of the RETS access I've seen to and MLS is NOT MLS-to-MLS exchange, but MLS-to-third party vendor, My gut reaction is that finishing out these remaining schemas (even as outbound-only transactions) is can of worms that many implementers won't be prepared to execute quickly, affordably, or safely. The return on risk seems pretty high to me.
Mike - I agree with you that the core property aspects are key and an obvious priority, and that some of the elements I mentioned have significant challenges to them. The point of this post was to look ahead ... what could be accomplished over the next ten years of RETS?
As one of my favorite poets said, "Honor the past and welcome the future."
It's awfully exciting, when we think about what standards could enable in our industry.
Matt Cohen has consulted to MLSs, Associations, franchises, brokerages, and many real estate industry software companies for over 12 years. Matt is a well-regarded real estate industry expert on industry trends, software design, product management, project management, and information security. Matt speaks at conferences, workshops and leadership retreats around the country on a wide variety of MLS-related topics.
RSS Blog Feed