3D Printering: The Problem Of Thingiverse

Most makers, I’m sure, enter into the 3D printing world with a goal in mind. Whether that’s printing enclosures for projects, Warhammer figurines, robot chassis, or even a mechanical computer, there is usually some obvious utility in having a 3D printer at home. 3D printers are a machine tool, though, and any time it’s not being used means it’s an investment with a lower return, or at the very least a really cool toy gathering dust.

Where then do you find new stuff to print that you don’t design yourself?

For the longest time now, Thingiverse has been the largest repository to share, browse, and download object other people have made. Even I have some very stupid stuff up on Thingiverse and have made use of a few random objects I found on there. This does not mean the 3D printer community particularly likes Thingiverse, however: Last year, Makerbot, the people behind Thingiverse, changed the terms of use so (allegedly) Thingiverse owns everything uploaded to their service. Couple this with completely unsubstantiated rumors of things being removed from Thingiverse that compete with Makerbot products, and you have a perfect storm of people unsatisfied with an online repository of 3D objects.

There is a huge market for an online repository of user-submitted 3D objects that isn’t controlled by Makerbot, and many have attempted to enter the fray. Defense Distributed, the guys behind the 3D printed AR lowers and all-plastic handguns launched DEFCAD, a Thingiverse clone, made an attempt by mirroring thousands of Thingiverse objects, removing the attribution in violation of these object’s licenses. Shady, yes, but at least it’s an option. There are other repos such as Cubehero and the newly launched YouMagine, a repo developed by Ultimaker. the Luke Skywalker to Makerbot’s Darth Vader.

But here is the problem with Thingiverse: even if you would like to get away from using this Makerbot service, it’s still the largest collection of 3D printed objects on the Internet. It has the most users, and is growing more each day than any of its competitors. Putting your objects anywhere else only means fewer people will see them, and fewer still will incorporate your designs into their new designs.

There are a few tools for you to ‘roll your own’ object repository. Github has a great new tool for viewing diffs between different versions of objects. There’s even a lot of work in making the Github landing page more like a Thingiverse page. This doesn’t address the core value of Thingiverse – if all the objects aren’t catalogued in one database, searchable by anyone, it’s just not as useful a site as Thingiverse.

I’m simply not smart enough to offer up a solution to this problem. Therefore, I’m turning it on to you: how should the 3D printer community retain the great value Thingiverse offers while still making something as usable as the now-malagined site? Should any new site mirror objects already on Thingiverse a la DEFCAD, only with proper attribution? Who should control the portal to all the objects, if anyone?

If you have any ideas on how to solve the problem of Thingiverse, drop a note in the comments.

75 thoughts on “3D Printering: The Problem Of Thingiverse

  1. yeggi.com correlates a bunch of 3D object databases making the searching simple. While this isn’t the best solution, it’s definitely better than just using thingiverse by itself since you can find the designs from those that left thingiverse as well.

    1. http://www.bld3r.com aims to find the best print-ready objects whether they’re hosted on our site or on others.’ So recently we’ve made a push to link to every site imaginable, including thingiverse, youmagine, cubehero, ponoko, github, instructables, grabcad, personal websites, universities, and museums like the Smithsonian and africanfossils.org. We’re also the first site to integrate Gary Hodson’s thingtracker protocol.

      As geeky as it sounds, we hope to be the reddit/ stackoverflow/ hackernews of 3D printed objects. As a lot of people are saying below, the web is the web, and loyalties to one website will come and go, but the web will always grow. So our idea was that makers should share what, where, and how they like.

      In response to “movements in the industry,” we’re dedicated to being independent so that people can find the coolest objects and so that builders can have complete control over their objects. We’re not a giant corporation, and we’re just two guys.

  2. Same could have been said about webpages. All you need are standard ways to share the metadata and search for objects. Everyone is trying to make the one object repository to rule them all, they have forgotten that the web is a web.

    1. I totally agree. My first thought when reading this article is : Why try to have more objects? You could have a host that is functionally similar to Thingiverse however it also has the additional capability of more or less searching other object sites like thingiverse and bringing all of them together in one useful list.

      1. Isn’t that just another version of the ‘one object repository to rule them all’?

        I think the parent was saying to use the power of the web itself. Just put your work on a web page. Link to other people’s web pages with objects on them, let others link to you. That’s what the ‘web’ is. We don’t have to give up ownership and control of or stuff by placing it all in some third parties unified database.

        I would mostly agree with this. But… I would also say, make those web pages easily indexable so that we can find stuff through search engines too.

        Maybe we would find that we need an object friendly search engine too. I don’t know. I suppose if there was such a thing it would probably accept referals of pages including objects that should be indexed. So.. it would be an index! But… it would be an index of people’s own pages, still owned and controled by the authors. It would still be the web rather than some html based single company owned content service.

        The problem of people forgetting that the web is a web isn’t just applicable to objects by the way. Look at how personal home pages have just about gone extinct, now it’s all owned by Facebook! The idea of an open internet where we all own our own ‘space’ is being replaced with a few companies to which we give all our data.

        Now I will hit submit and give this writing to Hackaday. :-)

    2. What if I don’t want to search for objects? I want to browse, preview, choose, leave reviews and also get updates about new content when it’s uploaded.
      People didn’t forget that web is a web. It’s just that web, although brilliant technical solution, is not really user friendly. Humans are always looking for simplification, so one go-to place for 3D models will always be preferred over open see of web plus Google’s search text field.
      As for solution – not sure. Alternative sites will need to somehow lure users/authors in. May be by paying authors per download from a pool of money generated by ads displayed on that web site? The more users, the more the authors earn. The more the authors earn, the more authors and content will appear on the site. win-win

    3. The comment about metadata is spot on. The problem, from my perspective, is that there are no universal metadata standards for objects. In my ideal world, there would be a rich set of metadata specific to objects along with the appropriate Dublin Core elements.

  3. This is exactly what iTunes did for the MP3/digital music ecosystem, IMO. I’d be surprised if the reason for the restrictive ToS on Thingiverse wasn’t more intended to prevent competition in the object catalog and less intended to screw contributors…

    Ask yourself, is there a decent alternative to iTunes out there? The problem is overcoming social momentum, and to do that I think you have to have a lot more service in your Service. You need to make it easier to use, and more empowering. This is how GitHub has become such a force.

    I’m a software engineer, and was one of the core developers of Apache Maven (java build tool). I don’t necessarily think we understood it at the time we designed Maven 2.0, but arguably the most valuable thing about Maven is its coordinate system for referencing other projects. That, and the repository of jars itself. These things abstract away a lot of detail about how to retrieve and incorporate other projects into the code you’re writing.

    If you want a suggestion for a new service, that’s an excellent place to start. Assign each object some sort of [intuitive] coordinate, and obviously a version. The versioning scheme needs to accommodate multiple “release” streams, not a single owner…that’s a shortcoming of things like Ruby gems, IMO. Anyway, once you have everything in a catalog assigned a coordinate, then the next step is to enable users to create sub-assemblies and compound objects by gluing these object coordinates together and sending them to a shop (which may be a single 3D printer or CNC+3D+…). It wouldn’t be that hard to namespace the coordinate systems so you could extend this build system into retrieving and combining software to run on the creation, too…Arduino libraries for instance.

    This is an idea I’ve been playing with for some time now, with a few friends. Unfortunately, we’re all relatively new dads, and between that and work, don’t have time to make it a reality.

  4. Is there a site like Thingiverse for Laser Cutters?
    I opted for a laser rather than a 3d printer, but I found that making SVG files is harder than it looked. So any Pre-made SVG’s that I can run through my laser may be fun to cut.

  5. Thingiverse does not claim ownership of its uploads. But what it does is a clear indication the lawyer drafting those terms did not understand open source at all (not really surprising from makerbot?). Basically the problem is this: if you extend/derive something with an open source license, most ofthe time you are bound by the terms of that license. So you do not have any extra rights to surrender to Makerbot by using their site. The uploader may submit to their terms, but all the other contributors can not.

    But bad as it is, this is just the network effect in action.

    1. If you designed your part in solidworks, then your part wouldn’t be bound by any open source license. Plus, very few licenses would affect other items produced using the open source tool, rather, they would only affect programs that integrated the open source components. It’s not like using GIMP or Open Office mean you lose any licensing ability for anything you draw or type.

      1. That wooshing sound?
        If you design a new part, then you own all the copyright to that part and you can submit to the extra clauses that MB put up. If you take a part and change it to your needs, you only own the copyright to the changes, not the whole part. So you cannot surrender the extra rights MB demands because you did not have them in the first place!
        Do note that open source is actually built upon copyright, the creators are just issuing extra rights (and conditions) with it so you don’t have to negotiate terms and conditions if you wish to put something online that was made by someone else.
        What program you used to design or change your part plays no role in that, or were you trolling?

  6. Good post, it seems like a lot of the existing and newer solutions out there are lacking one feature or another. It would be nice to see a standalone entity come up with a 3D objects database with a really intelligent backend system. These ideas are just for 3D printing oriented objects, I’m well aware that Thingiverse also has CNC, laser cut and other objects as well.

    It would be really nice to see some kind of a rating system for objects other than ‘likes’. Something more in depth such as rating printability, quality and some other design variables. This could be a scaled rating system based on simple variables such as the age of the user, the number of the users uploads, or ratings on their own objects from other people. If User2 rates someone’s object quality as low, but his own objects receive low quality ratings, his quality rating on User1’s object would not hold as much weight as someone who has received high quality ratings on their objects.

    If the system could find broken or unprintable files/models (STL files for example) and also analyze the quality/resolution of the output files, maybe do some curve analysis of the STL and determine how many segments make up the radius, etc and generate a Model Smoothness factor.

    I see more exact copies of objects lately without attribution or being marked as a fork/derivative of another objects. I’m not suggesting there be some kind of a DRM but it would be cool if the system could tell the uploader ‘Your model seems a lot like xxx model, is it associated and would you like to list this new item as a derivative?’ and if it’s a derivative ask them to enter the changes from the original model.

    I think that covers my three biggest problems with digging through 3D models. 1) A good rating system with weighted ratings based on experience/participation/skills of the raters. 2) A system capable of analyzing files and notifying the content creator and possible downloader that the files are corrupt, unprintable or may be an issue. 3) Duplicate objects with no obvious difference in design, or stated in the description. Oh yeah, 4) if multiple people with a good track record report an object is not printable, flag it as having possible issues and notify the creator.

    1. I would add a 4th item to a list of problems:

      Some models can only be printed on a machine with certain features, (say dual extruders for filler material), or on a machine with a certain capacity.
      How about a list of machines that successfully printed a given model? Or a list of capacity requirements of a 3d model? That can be generated with software that can parse an stl file, I would think.

      And if designed but not successfully printed yet, nothing wrong with that, just say so.

  7. I feel dirty using Thingiverse, but the crappy search feature and the instructions, comments, and lineage for each thing are incredibly useful. Nothing else offers all of these.

  8. I would find it useful to have an ignore option. Because printing is still developing, 80% of new designs uploaded are printer mods. People finding something that helps their own machine work. But once you have your own one working nicely, you really don’t want to hear about every minor alteration to every type of machine available. How many spool holders can you need?

    14% of designs are completely irrelevant to anyone except the designer. An air conditioning vent slat for the left hand side dashboard of a 1989 . Or designs of someone who is bored and wants to design for the sake of it, rather than actual usefulness. .

    5% are kiddies who have never seen a printer but discovered how to convert a file they downloaded from Sketchup Warehouse. Unprintable.

    It would be pleasant to avoid mining through all this space filler to reach the 1% of truely clever/useful/novel/worthwhile golden nuggets.

  9. I would love to see The Pirate Bay get involved. Although their interface isn’t designed for this, their infrastructure is perfect! If they put up a custom interface for a section that was dedicated to “objects”… printing, scanning, milling, cutting, etc they could corner the market quickly.

    If not Pirate Bay, I think RepRap.org , Wikipedia, or CNCZone.com should step into this role. Defining it as a goal, and then asking for volunteer developers would probably be enough to make it happen. Any of these four sites has the “cred” to pull it off and be trusted by the community.

      1. Yes, I’m aware of that. My point was that their interface sucks for this purpose. Navigating by individual file is a bad idea. Navigating by “Project” would work a lot better. Individual files kept within the project page (using versioning) would be the actual torrents. There are a ton of improvements to the interface that would be necessary to be competitive. Things like having unlimited images on the page, threaded comments, being able to embed video tutorials, related projects, better search features, etc…

        It’s all about the interface. The Pirate Bay has the necessary distributed infrastructure and the trust of the community, but this is a specialized purpose and it requires a specialized interface.

        1. Ok, wasn’t sure if you knew. I agree with the rest. From a quick glance, it seems like they’re providing the section mainly to provide a means of accessing controversial objects that may not find reliable centralized hosting, not trying to be a serious contender as an general object repository.

    1. ThePirateBay IS the perfect place for this!
      I don’t understand why people try to make so slightly different copies of something that already works. Use ThePirateBay to store everything in the torrent part of the world and the only thing left to build is a.design a cuter interface for it directly on TPB or b.design an indexer website to filter through the physibles (like thingiverse) but link it to the tpb page!

  10. http://xkcd.com/927/
    I agree with many commenters above. I also think that what is needed is a good thing search engine for all the thing sites. To do it right, there needs to be an [xml|json|etc] interface provided by thing hosters, this is like the begining of podcasting, OPML, and RSS combined. ThingTracker (mentioned above) looks like a good start. These other thing sites should take a look. Is anyone using thingtracker’s Spec yet?

  11. someone start a kickstarter campaign and we all chip in to buy thingiverse and give it to hackaday or the instructables part of 123d now, or just elect a team of citizen hackers to run it… i mean, everything’s for sale right? if you get enough money… hahaha.

    1. Except the claimed advantage of Thingiverse isn’t actually true.

      >This doesn’t address the core value of Thingiverse – if all the objects aren’t catalogued in one database, searchable by anyone, it’s just not as useful a site as Thingiverse.

      Thingiverse has a very wide-ranging set of forbidden objects/topics that keeps thousands of designs off its database. It may be the current largest in terms of upload number, but it is explicitly NOT the wikipedia of home fabrication. The longer people treat it like it is the more it hurts us in the long run. Bite the bullet and stop uploading your designs to it.

      1. and where do i post my all reprap design ? if not on thingiverse ? If i read here, i find at least 10 places not finished,in beta without user.. but which one ? a part thingiverse right now ?

        1. @serge I saw your question about where to post. I’m running Cubehero (https://cubehero.com). And like the article above said, you can try to check us out. I’m always free and open to feedback and your concerns: wil@cubehero.com.

          For Cubehero, my focus is on making it easy for people to work together on 3D printed projects, rather than to amass a large collection of random things. I believe with the democratization of manufacturing tools and the distribution of the internet, it’s possible to make things that just weren’t possible before–things that change the way we work, live, and play.

  12. Hey hackaday.com’s new owners… the solution is right in front of you. Where do most makers and 3d geeks go for kick ass projects?…. yeah… here… you paid good money for hackaday.com, why not start a re-branding effort and add a component to the site that allows a social integration for makers to upload their projects and host files that are attached to their project? Feel free to send me royalties from this!

  13. GNU Mediagoblin – a decentralized media hosting web framework – has support for stl and obj model uploads, specifically intended to be used as a thingiverse replacement. Its a little rough around the edges, but Mediagoblin has a pretty active community of contributors, so its likely if it saw more use, it would see more development.

    Mediagoblin was originally intended to be a decentralized, federated replacement for flickr and youtube, much as pump.io and status.net are federated alternatives to twitter.

    This old news post announced and explains the feature mentioned above: http://mediagoblin.org/news/3d-support.html

  14. I realize everyone would love to have an all-in-one solution, but I think in there may actually be a better approach in serving specific niches; I believe first and foremost sites for collaboration and forking designs needs it’s own place. I think you’re in a different mode as a user when you’re doing that. Then, on the flip side when you need a design of a certain type (say automotive parts or your favorite movie or game characters) you go to a site focused on being quality parts for that use (because really, you just want to get an item and print, not change Yoda’s ears).

    Especially when dealing with the issue of quality, printable items, I think focused sites may best solve the immediate problem. I realize you can find just about anything on Craigslist, but think about all the sites that have built great experiences off of single categories from Craigslist: http://cdn.theatlantic.com/static/mt/assets/science/HannahPost02.jpg

  15. I really hate browsing all these sites… they’re peppered with random shit. some things are useful, some are toys, some are artistic, some are tests…. I feel this is because they all have to be BROADLY categorized. I’d rather see more user groups and categorites – perhaps user meta-tagging or tree views of things. If i want to print a phone case, i dont want to look at ALL CASES for ALL PHONES, just the phone I HAVE, right? Same with 3D printer parts., or figures, or whatever….

  16. I haven’t been at this long but if thingiverse is eliminating the competitors objects, they certainly seem to have a plethora of Ultimaker pieces and modifications, and a search for reprap give 3685 results. I haven’t started any personal projects there yet but when I do I certainly will study in detail the terms of use, nobody’s getting the exclusive on distributing a project I start.
    I appreciate everyone’s input on this issue, the comments have been most useful.

  17. Prusa helped spreading the paranoia against Makerbot in Thingiverse. Thingiverse needs to protect itself from morons suing Makerbot for hosting their property and showing it in the frontpage, the terms of use shouldn’t be surprising. Also, I don’t know about you guys but I’m not posting anything to Thingiverse I’d like to keep for myself, I post things I want to share and I love to see other people making derivatives of them. Creating Thingiverse clones is OK but it just introduces noise.

    1. the terms of use are surprising, as they clash with something MB was known to promote: Open Source. Their terms and conditions demand extra rights from the users which is impossible when you did not make your own original part, but adapted an existing (Open Source) design to your needs. Then the only rights are what is stated in the license, not negotiable as their lawyer seems to have thunk.

  18. problems with thingyverse:
    the terms are considered unfriendly by some but are imposed by makerbot
    vast categories of objects are disallowed by makerbot
    Makerbot apparantly will delete objects that compete with it
    it is certainly true that the automated build platform that makerbot copyrighted is very close to existing rought designs, and even completed working models uploaded to the thingyverse.

    problems to launching a thingyverse competitor:
    thingiverse has most users, and it is expected that most will not leave there
    this user base grows each day.

    buy thingyverse. change terms to how you want them.

  19. Item repositories really need to be run separately from any maker of printers. Even open groups like reprap, let alone profit oriented companies, will have incentives to make the repo work best for their system.

    As for the problem of empty repos not drawing users, that is just hard work of creating new content that any site has. Don’t just stuff your database with copies, make new ones that show off the features that make your repo better, and convince some other people to try it out. Then, if the stars align, others will switch over until it avalanches.

    1. YouMagine from Ultimaker is actually ran mostly separately from the printer fabrication. (people working on the printer are different people then the website people) It has an open API, and it could split off from Ultimaker.
      We have made a coupling from Cura into YouMagine. But Cura is not limited to Ultimaker printers (and it actually gaining a lot of ground with the Delta-bot people, as default Ultimaker settings seem to work very well on delta-bots)

  20. Thingiverse sux. I could’t even use a search engine with Opera mini running on my tablet. That website just doesn’t work. No link to desktop version.
    Also this scroll-auto-load makes me mad, a lot of playing with mouse-wheel up’n’down and wasting time, memory, cpu and internet traffic which is quite expensive when you’re on $/kb plan. Especially if page loads all the pictures you don’t even want to see.
    Where the hell is link to particular page number and how to get to the top menu? Why that website reload everything over’n over again all the time? How to disable this script?
    Does any1 know?

  21. All the Makerbot/Make/community/open-source(over-used term) bullshit is getting out of hand.

    It’s a fabricated illusion for some kind of fake rainbows&unicorns capitalism to profit from a half-assed democratization of technology, without real world knowledge of limitations and capabilities. Communications experts at giving a false sense of community using meaningless associations to build a customer base.

    ex: Who the hell uses solid core copper wires connecting a mobile part?!?!
    answer: Someone that doesn’t know shit, and doesn’t care about knowing electricals.

    ex: Who the hell puts the washer on the opposite side of the nut, round face down,
    and proceeds to grind the nut into the case?!?!
    answer: Someone who hasn’t the slightest clue and passion about mechanicals.

    I expected more from Bre, but I’ve always been warned about ‘ideas guys’.

    So this article is in no way surprising to me. More to come I assume.

    And yeah I may be a damned punk, but I can smell ‘The Man’ and his Grandfalloon miles away. Shame for the vitriol but I’ve seen so many good guys with faith burning up in rage.
    Just ask EEV’s Dave about his ‘bot…

  22. I would better work on a standard “tag” system for a web page having 3d model to print. Soon , google will integrate it in the search engine, then thingiverse and all others will become .. “soo 2012” :)

  23. Please don’t propagate the line that “Thingiverse owns everything uploaded to their service”. It’s easy enough to go read the TOS and find out that in reality Thingiverse makes make no claims to things uploaded by users. What they claim is that by uploading things to the site you grant them the right to serve the things to other users. Because if you didn’t grant them that right, it would be illegal for them to host the files or to distribute your things to other people according to US law.

    So there’s no “perfect storm” of people disliking Thingiverse. There are a few people who don’t like the idea that the only big repository of 3D printable things is run by one of the 3D printer vendors.. But Thingiverse’s usage and content collection and functionality are growing, not shrinking, and none of the proposed competitors have any traction, so there’s no sign that there’s any actual movement away from Thingiverse. And I think that it’s not terribly likely that anyone is going to be willing to fund the development, hosting and ongoing operations of a site to compete with Thingiverse on better terms than MBI, who runs the site for free, publishes all of the site APIs so anyone can integrate to it, and lets anyone use it for free, with minimal restrictions. Specifically, since the site’s target is to promote 3D printing to the general public, and specifically as a place where teachers can send kids, they allow nothing illegal, no real weapons, and no porn. And it’s run by a separate team of people, operationally independent. And it’s pretty well funded, with good production values.

    That being said, if someone else wants to spend all that time and money to build another Thingiverse, do it as professionally, and to do without the sponsorship of a 3D printer company, and give it away for free, that’d be nice. But I don’t see that as particularly likely. So far they all are either tied to another printer company, or charge money, or are an attempt to DRM lock content to force people to buy designs to print. So not better than Thingiverse, which is part of why none of them have any real traction.

  24. wasnt there a program that could take an offensive object say like a gun and encrypt it into a 3d smear or looks like a 3d ocean wave looking hair on the head of terkina from disney tarzan.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.