Best Practices for Publishing Prints  

RE: Best Practices for Publishing Prints

I'm sure other can add to this list.

Actually, sounds very comprehensive to me. 

RE: Best Practices for Publishing Prints

Same here.  I might borrow this and give credit of course.  

RE: Best Practices for Publishing Prints

After thinking about this a bit more, I thought of a couple of other things I think are worth adding :

Photos -

• Of course you want a photo of a completed print, both by itself and "in use" (if that is applicable).
• More photos is better (within reason).
• Include something in the photos for scale.
• Photos showing recommended orientation on the print bed are often useful.

Discussion of the print and print instructions -

• It's often useful to explain what worked and why, but can be equally important to explain failed attempts and why they failed.

Model name -

• The more descriptive the name, the better - even if lengthy and redundant (helps with search)
  • For example :
    • Poor : Animal
    • Better : Cat Sleeping
  • Include as many relevant tags as you can thing of - feline, pet, lying down, etc.

RE: Best Practices for Publishing Prints

Agreed, photos are important, I guess I just took it as a given but it's worth spelling out. At least that's an indication that the model is not just a stupid student CAD design assignment (all too common on Thingiverse) but something someone has actually printed—and is printable!

RE: Best Practices for Publishing Prints

If including a Prusa Slicer Project as the 3mf then the need to show placement orientation is not so needed as when doing that you can place the part, add support enforcers, paint the object etc all for the end user.  In inherent in the project.   Of course that doesn't help them if they are using a different slicer.  

I would still argue that including the stl files is best if you want the largest compatibility as not everything supports 3mf yet.  It all depends on your 'target' audience.  

RE: Best Practices for Publishing Prints

Neophyl - you make a good point, but I still think photos of the actual print-as-printed are worth including, as the views you get in the slicer are somewhat idealized.

Also - just a guess on my part, but I suspect that all major slicers in use today support 3mf, and STLs can be generated from 3mf if really needed (and other ways to do it besides using a slicer).

If we're talking *best* practice:

TEST the part properly, use it for a significant period and address any failure/wear patterns.

If the print requires non-trivial post-printing assembly: TEST and refine the assembly instructions.  At the very least least print one and observe a friend attempting to follow the written instructions.

Cheerio,

RE: Best Practices for Publishing Prints

Are there any common slicers that will not both read and import from standard .3mf files?

The reason I am asking is that occasionally someone will post a .f3d file and no .stl or anything that most slicers can read.  (I've seen this in the local Printables side.)

Personally, I prefer individual .stl files over .3mf, and particularly over .gcode files only.

I would also suggest running all .stl files through any of various stl validators.  Thingiverse, in particular, is notorious for having quite a few wonky .stl files.  (Yes, I admit that PrusaSlicer tends to not be as forgiving on these than Cura and some other slicers.)

RE: Best Practices for Publishing Prints

I used to use the one that came with the Davinci printer (as it was/is proprietary) and that would accept stl and amf files.  Not 3mf.  Not sure if that's still the case as I moved on from that printer a few years ago.

While we may hate it, the lowest common denominator still is the stl file.  If you want the widest range then it is unfortunately still the 'go to' file format.  

RE: Best Practices for Publishing Prints

STL or 3mf, it's not an either or. I usually upload the STL and if the model requires specific modifiers, placement, supports, etc, I'll post the 3mf in addition to the STL. And why not. Redundancy is a good thing, and the size of these files is so tiny that in this day and age of storage technology I don't think anyone has to be concerned about wasting disk space. 

RE: Best Practices for Publishing Prints

Diem - very good point about testing. This is another reason why I like to see a photo of the print as actually printed. I do make an effort to test all my prints in actual use. Even then I may sometimes find a problem after using a print for some time, in which case I think is it good form to add an update to the original posting noting any issues or problems. Conversely, I sometimes publish a print with known problems (and I note this when I publish), with the idea that someone else may find the basic idea useful and be able to improve it.

fuchsr - my concern with adding multiple formats when publishing is not a concern about disk space, but more about keeping track. When I am creating a print that I think I am likely to publish, I typically create a non-published "stub" and then add to it as I go along. I find that this method is a good way to keep relevant notes and files collected as I go along. Suppose, for example, that I am creating a new print with several parts. Each of the parts will likely go through multiple iterations. All of the iterations are saved to my desktop, but I often upload the "final" version to the stub. Sometimes, I find that as I create the additional parts, and earlier "final" print is not so final after all, so I have to delete all previous versions from the stub and upload corrections. Of course I could just wait until everything is truly final, but then I risk leaving something out, and losing track of details worth noting. In addition, I typically take a lot more photos than I publish in an effort to get a few good useful ones; it's easier to sort through those immediately than to wait for completion of the entire print. I also find that this piecemeal approach is much easier to put together as opposed to waiting until the end and composing the entire thing from scratch. This method works for me, but YMMV.

RE: Best Practices for Publishing Prints

While we may hate it, the lowest common denominator still is the stl file.

Er, highest common factor?  G,D&R...

STEP is the industry standard and Prusa Sliver does accept it.  Most of industry can handle .stl files but they don't like the lack of defined scale data.  .3mf is a bit of an oddity in that it addresses the major drawbacks of .stl but surprisingly few people in industry have yet encountered it.

Right now .stl, for all it's shortcomings, is the clear leader in hobby/home printing but I think most have at least heard of .3mf.

So it really depends what you think your main customer base is.

Cheerio,

RE: Best Practices for Publishing Prints

In thinking over the issue of 3mf vs stl, or both, I am now inclined to lean towards including both as (so far as I know) there is some slicer software out there which does not recognize 3mf. This is a point I should have recognized earlier, but as I use PrusaSlicer I tend to forget about the rest of the slicer universe. Of course I could just take the position that anyone can get PrusaSlicer or Cura or whatever and do the 3mf to stl conversion themselves, but why not make life easier for others and make the stl available as it is a minor extra step for me.

In addition - a small amendment to the conversation on testing prints. I still agree with the point that a print should have been actually printed before it is published, and tested if feasible.  I make the additional condition because I recently published a "snow scoop" for clearing snow from a snow-thrower chute.  I can't actually test the print in use until we get some snow where I live. If I waited to test the print, it would not be available to anyone else who needs it, so I went ahead and published it in case anyone else want to try it out with that limitation in mind.  In short, I think it is OK to published an untested-in-actual-use print so long as that fact is noted.

I still agree with the point that a print should have been actually printed before it is published, and tested if feasible.  I make the additional condition because I recently published a "snow scoop" for clearing snow from a snow-thrower chute.  I can't actually test the print in use until we get some snow where I live. If I waited to test the print, it would not be available to anyone else who needs it, so I went ahead and published it in case

But it's untested and therefore unwarrantable.  If a project takes a year to test then it's not worth publishing until it has passed that test - which might take several iterations, several years.  Until that point you are merely wasting other people's filament.

There is nothing more irritating than downloading a 'work in progress' print to find that only the easy first 10% has been done.

Another horror is the posting of every part of a multipart project as a seperate print.  One project should result in a finished object, not a part that is useless without searching out and downloading fifteen others.

For example, we see founts presented as 26 or 52 individually published prints ... why not one 52 part set?

We see multiple minute variations of one basic idea ... why not one 40 part compilation from which the interested user can download the parts they want without having to trawl the database?

Cheerio,

RE: Best Practices for Publishing Prints

Seriously, is there any reason to publish a print before (or without) actually printing it and at least giving it some kind of evaluation as to if it will perform as expected?

I have only published a few prints here, but each one was indeed actually printed and in use as intended.  Plus, I've only uploaded prints that I thought others might be interested in, and, of course, nothing trivial.

I hate to keep circling back to this sore spot, but as I now think about it, I'm sure that big core-dump of prints that came in right after the 'competition' started included some which had not been tested (or printed).

As to why some people break up obvious sets, not to keep with this broken record, but by splitting a font into two 26 character sets, it's counted as two uploads and two downloads.

RE: Best Practices for Publishing Prints

I've noticed Printables is attracting a more and more international public lately, which is a good thing...
But I've also noticed a more frequent use of foreign languages, which make it hard to search for items.

RE: Best Practices for Publishing Prints

Seriously, is there any reason to publish a print before (or without) actually printing it and at least giving it some kind of evaluation as to if it will perform as expected?

Of course you knew the answer! 😉 It's the morons who are too dumb to understand how the Prusameter reward system works. The ones who think uploading crap gets them free stuff and don't realize the only way too get serious value out of the system is to publish quality models that people actually value so you get hundreds of downloads per month.  

And as you also pointed out correctly, these quick competitions or whatever they're called certainly don't help either. If you prefer quality over quantity, one of the worst ideas ever. 

RE: Best Practices for Publishing Prints

Posted by: @yveske

I've noticed Printables is attracting a more and more international public lately, which is a good thing...
But I've also noticed a more frequent use of foreign languages, which make it hard to search for items.

This is true.  It is too bad they cannot include an algorithm in the search to translate.  

RE: Best Practices for Publishing Prints

I agree with that an actual print of a part should be made before publishing, yet I can still think of exceptions to the "rule." For example, suppose I have a part I want to print, but have not been able to print successfully - can I not publish that model and ask for advice on getting it successfully printed? I have never done that, but it seems reasonable to me that someone might do that so long as they warn people ahead of time.

I definitely do not like it when a single print composed of multiple parts is published as multiple separate pieces - just  seems like more work for everybody.

On the other hand, there are understandable reasons for publishing "sets" as multiple publications. I have published a print, and then in response to feedback published one or more additional related prints. It's not because I want to drive up my numbers (most of my stuff was published before the Prusa "reward" system went into effect), but rather because there are sometimes benefits to me and other in doing it this way. For one thing, the publishing system does not inherently do versioning - this makes it inconvenient and messy to add additional versions to an already published print.

Let's take the alphabet example :

• Scenario one : I publish a print of the letter "A". Nobody is interested. Why bother to publish B thru Z?
• Scenario two : I publish a print of the letter "A" and many people like it. It takes me a week to design and test print a letter. Should I make people wait another 25 weeks before I publish again?

The testing issue is also not straightforward. I do make an effort to test my prints, but what works for me may not work for you, so arguably a print has to published and used by others to thoroughly test it. This is part of the value of publishing - it enables makers to get feedback and make improvements.

Ultimately I think the real solution is truth in advertising : When you publish a print explain whether and how long you have tested it. Explain both the negatives and the positives about the print.