Wednesday, January 21, 2009

Paying for Families

As the famous saying goes, "You get what you pay for".

We obviously don't want to pay for families, but one way or another, we're shelling out money to scour the web to find good content or to build our own. Whether you're hiring someone to build families for you or buying them ready-made, it costs time to communicate intent and check the work. Need I say more?

By now you might have heard about Turbosquid for Revit. As the description on the site goes, "Revit Market is designed specifically for Revit users to buy and sell Revit families, scenes, and visualization content." So if you have any families you would like to sell, I encourage you to post them. Or if you're short on time, maybe you'll find what you want at a reasonable cost. You never know, it might even cost you less at the end!


If you read my long posts about BIM content a while back, you know how I feel about the subject. Content is a complex topic, especially when it comes to leveraging BIM throughout the entire life-cycle of a project. I still think that ultimately, the companies that manufacture building products need to provide this content themselves. Only then can the design team trust the accuracy of the data. And would we pay for it? I don't think they'd dare to charge us ;) I believe competition would take care of the situation. The first manufacturer of a particular product that doesn't charge gets specified the most. And the rest will be history.

Oh, one more thing. Until the end of January, you have the chance to win some cash at Turbosquid. So if you have some spare time, now you have an assignment on your to-do list and you might end up a little richer in the process!



craig barbieri said...

I love the idea of selling families that I have made but hate the idea of having to buy them - That's what everybody's thinking to themselves right?
Is this a good business model? I don't thinki it is [for Autodesk]... look at Google Sketchup for example, the library of models is huge because it's free. Sketchup is popular because it's free unless you use it professionally(and it's easy to use). The Autodesk Seek library is Autodesk's response to the google library and I think it will be the main source of content if more manufacturers continue to join its ranks.
Turbosquid is a nice idea for the owners, but unrealistic if they expect to become a major content library/source. I can see it evolve into a library of very complex, detailed, hard-to-build, advanced content, but if they expect to extract themselves a huge library of this content from user's submissions to their "contest", then I don't know how to respond. Is that a tad bit unethical? No, but maybe questionable.
What keeps Joe-Intern from grabbing all the Component families in the firm he's working at and uploading them for sale to his profit? Or someone else who somehow got ahold of a project model of and posted components for sale that didn't belong to them.
Oh, and the other thing is, if they are going to charge for these components, they should get real... prices are ridiculous - half of that stuff looks like it's old stuff from the free library at RevitCity!
Personally, I think that Components should remain mainly free and foster the collaborative trend that is growing in our industry. Fees for the components that are next to impossible to build by 95% of users are justified but windows, doors, furniture, MEP components? - these should come from manufacturers or in-house.

Dave Baldacchino said...

Hey Craig, thanks a lot for your detailed comments. You raise a lot of good points, especially when it comes to intellectual property of families. We have no way of signing them or restricting their use (such as posting them for profit!). I mean yeah, we can sign them, but we all know that's a joke and so easy to remove.

If I created a family and was paid for it, it would be dishonest of me to upload it for profit, unless the person that paid for it agrees and perhaps gets a cut of any potential revenue. But what would the dishonest person do and how do you control it? The best systems on the web are those that self-police, or at least do so with minimal intervention.

Building content is not that difficult. Building effective, functional, parametric content is not easy or cheap. I'm a bit skeptical as to where Turbosquid will end up. It will be interesting to see!

GeoffB said...

“The first manufacturer of a particular product that doesn't charge gets specified the most. And the rest will be history.“

While this may indeed be true, any architect who does specify building components based on the availability of RFAs (or even DWGs or PDFs) is skating across thin ice from an ethical stand point. We are duty bound to recommend the best products for our client’s needs, not simply those that best integrate into our design process. In the end it’s building performance that matters, not software performance.

Specifying a window because there is a nice family is not that different than specifying it because the manufacturer pays you a kick-back to do so. In either case it represents a conflict of interest by any professional who’s role (contract) is to advance the interests of the owner.

On the other hand, if these competing interests are on the table such that the owner can participate in the decision making process then I feel the conflict disappears. If the owner chooses Window B over Window A because that means they get their design faster/cheaper even if Window A performs better (or is cheaper, better looking, etc.), then so be it. Of course that would be a stupid move, a dollar chasing a dime. But at that point it’s the owner’s mistake to make.

Dave Baldacchino said...

Thanks Geoff, you made very good points.

More than specifying a particular product based on the availability of a family/other content, I really meant to say that one would find it easier to use as a design basis. In public project types, we cannot even specify just a particular product, but we use one as design basis and then list acceptable manufacturers.

GeoffB said...

Hi Dave, sorry for taking it a bit off topic, but it’s all realted.

Using families as placeholders in initial design or public projects is an important subject. How generic can or should they be? How much work is involved in swapping in the real thing? What are the consequences of not knowing the exact components to the design and detailing process?

But using a family from a manufacturer as a placeholder is not specifying and we need to be careful to make the distinction, especially if the component is iconic or has a lot of embedded identity data.

Dave Baldacchino said...

It's all related, so no worries about commenting. I actually appreciate having reader feedback!

We usually select the most common/widely installed object and use it as design basis. Usually variations aren't THAT bad if something else gets selected, but I think that's why it's so important for the team that designed and produced the documents to also be involved in construction. Swopping with "the real thing" would typically just involve selecting all instances of a particular family and changing to another one. The imortant thing though is that they are built the same (same centerlines, elevation heights, etc.), which I think is the hardest thing for us to control. This process might mean that some references are lost, such as dimensions for example. I've personally never done this as contractors are responsible for As-Built drawings in the projects I worked on.

As to detailing, you just have to focus on what's really important and know where changes can occur. So for example if I don't know exactly whether a mullion is 7 1/2" or 8 3/16", then I detail such that if it "grows", it won't affect anything else. So I dimension to communicate what matters to me the most (say, the exterior face) and that's it. It's very important that whatever we deatil isn't soooo "exact" that we get a bust every time a dimension changes. Typically, we're not building space shuttles, so we need to detail with building tolerances in mind.

Post a Comment