Discussion: FreeCAD is not ready for 1.0

[triplus]

Yeah, i propose that both camps first blow off some steam:

https://forum.freecadweb.org/viewtopic.php?f=8&t=43484
https://forum.freecadweb.org/viewtopic.php?f=8&t=43461

P.S. Around March we can start to conduct a more serious discussion.

[Zolko]

https://forum.freecadweb.org/viewtopic.php?f=8&t=43461

that's a circular dependency

[sgrogan]

An important technical note, to me at least, is semantic versioning
I think source code management and dependency management will be more complicated after a 1.0 release. The issues can obviously can be overcome, but I don't think they should be ignored.

As for using the year, I think this a branding/marketing thing, most/all softwares I've seen that use this have an underlying semantic version. Maybe going this way avoids my concerns, as the underlying version will respect the semantic versioning convention?

[frabice]

I think it really depends on what you use CAD to do.

If you work somewhere and your job is to design a load of parts that will fit together. Assemble them in CAD and check nothing clashes. Make a technical drawing including lots of hole call-outs, properly speccing tollerances, parallelism, etc. for external workshops. And then needing to modify parts to improve performance. This is a very common usage of CAD and it is not there for this.

Part design hole wizard regularly gives the wrong sizes. The hole information (such as thread and depth) doesn't make it through to TechDraw. Many thing that a professional workshop would need specifed in a technical drawing are not possible. It is unclear which assembly tool to use and how long it will be maintained as they are not part of core. Then when you come to do iteration two you get hit by toponaming issues.

This is not to say FreeCAD is bad. It is wonderful. TechDraw is improving at lightning speed, part design does most of what one needs except reliable holes, and hopefully Toponaming is more robust in realthunder's branch. When I worked at NIST (part of the US government) I used FreeCAD because I was frustrated with learning new CAD systems as I moved between groups, but I regularly hit a wall and had to send STEP files to others to make technical drawings.

This argument of "it is v1.0 for my workflow" is pretty meaningless. It would be good to talk to people in MechEng departments and see what features they would need. Sure some are in love with one program, but many people have used multiple programs. To shut down all criticism as "they don't want to learn our workflow" is not going to convince people to engage if you wont engage with them.

For this argument of v0.20 v v1.0 it makes sense to look at something like Inkscape which is used by loads of scientists to create their figures for publication even though many have access to proprietary software. Look at all the papers from LIGO. Inkscape are still not 1.0 because they have a clear roadmap and they will be 1.0 when they reach certain goals.

If the topic of v1.0 keeps coming up perhaps it is time to have a roadmap? To seriously discuss the pain points that come up often and ask if they are a "wont fix" because our flow is different, or if they are things which have workarounds but the feature/fix should be on the roadmap?

I am a very new Freecad user and I have been reading the forum for some weeks. You summed up my thoughts very well. Thank you!

[vocx]

Hi @vocx

I am prepared to discuss this, as such discussion would likely happen in March anyway. But if this is the thread we will have such discussion please change the title to be more neutral.

I don't understand what you mean. Neutral in what sense? The thread title was picked with the purpose of explaining why FreeCAD is not ready for a 1.0 version, which I explained in the first post.

[vocx]

...
- And then, you could estimate a time needed for this essential minimum requirements to be implemented or solved.

This is basically what I outlined in the first post. Which is why I think a 1.0, full featured, stable version should come in about 5 years. Implementing something big like App Link took about two years of development. Now it's merged, but we are still testing it. Testing it inside the master branch should take another year at least while the Assembly3 and Assembly4 workbenches mature and become more widely used.

[vocx]

...
I am tyred of telling people "No FreeCAD is not beta for what I would like to use it and to show it to you. ...

I dislike this attitude a lot. It basically means "it works for me, screw the rest". That's a very egocentric approach. What ever happened to being part of a community? I don't want a single part of FreeCAD to succeed; I want all parts to improve and be great in the same way. Sitting in one corner and saying, "we are fine, we don't need to do anything else because this works for me", just isn't the right attitude.

It is unquestioned that many users complain about FreeCAD being sort of disconnected, like each component works in their own particular way. That's completely true. The reason is we haven't made a concerted effort to improve the integration of all workbenches.

This was brought up again in this thread. Why not deactivate the Draft grid when we enter the Sketcher edit mode? It's a logical step to make both workbenches better.

...
But why only a checkbox for Sketcher and Draft grid, FreeCAD is modular by design, there can be more grids:

Because Draft is a standard workbench provided with FreeCAD by default. The Draft grid is a "core" component of FreeCAD, so to speak. Other grids would be secondary, as long as they aren't integrated into the master branch. (It makes sense to further investigate the grid in Chris_G's post above.)

And yes, FreeCAD is modular, but we should also make the workbenches more aware of each other so that the overall experience is better.

One criticism of FreeCAD is that it is seemingly composed of separate pieces, and doesn't look very well planned with a single user interface in mind... and that's exactly the truth! Wondering about the overall UI

We should definitely try to integrate the pieces better so that the overall experience is more uniform. This is a good thing.

No, i don't feel adding a Draft grid checkbox in Sketcher task panel is a good idea...

It is not well received. Integration is a bad thing? Why?

[vocx]

voxc would be good if you would read.some of this one https://forum.freecadweb.org/viewtopic.php?f=8&t=34923

You must be joking! I explicitly mentioned that thread in the first post of this thread. I have read that thread, of course.

For me this topic is all about version numbers. I do not care about features in this topic. For me we gone have enough features even for a version 2 or 3 or 10. I just would like to get a higher version number!

And in all fairness, you weren't able to convey your intentions in that thread and it was hijacked.

Your thread was only about the version number, not about stability or features of the software. So, in that sense your thread changed to a wider discussion. This thread is more about the features, and not so much about the number. I did mention my preference to just make a jump to 1.0.

[vocx]

https://forum.freecadweb.org/viewtopic. ... 20#p294666

... I just would like to get a higher version number!

...
A version number above 1. or at least near 1. is much better to sell to other people...

Again, this to me sounds like a very self-centered view. For you it works perfectly, but what about others? Do you even think about others? Do you only think about IFC and FEM? Why don't you try expanding your views to consider the rest of the users, or workbenches? Or the user documentation? Or the programming documentation? Or helping out with something that is not FEM?

You are actively trying to sell people the idea of a more mature FreeCAD, and people don't accept it because of the version number. That's a pretty odd thing in my opinion; it's basically inflating grades.

I want to sell the software on the quality of it, not just on that number. If companies don't want to use a less than 1.0 version, why does it matter to you? Why does it matter that somebody else spends thousands of dollars or euros in another proprietary software? Does it impact you directly? It's their money. Just like Linux, those of us who know about its power, use it for free, and we are happy about it; we cannot force people who use other paid systems to switch just because.

If you were committed to that cause, you would fork FreeCAD, remove all workbenches, except for maybe Part, FEM and Draft/Arch for IFC needs. Then you would call that BerndCAD, and say it's version 1.0. Does that not solve your problem? You would just copy the entire FreeCAD source code, and repackage it with a new version. Isn't that the idea behind LGPL2? Everybody can do it.

[vocx]

...
This argument of "it is v1.0 for my workflow" is pretty meaningless. It would be good to talk to people in MechEng departments and see what features they would need. Sure some are in love with one program, but many people have used multiple programs. To shut down all criticism as "they don't want to learn our workflow" is not going to convince people to engage if you wont engage with them.

Yes, I agree with that sentiment. Unfortunately, FreeCAD is still pretty small as an organization (literally no organization), so this will only come as more people use the software and decide to participate in its development.

For this argument of v0.20 v v1.0 it makes sense to look at something like Inkscape ... Inkscape are still not 1.0 because they have a clear roadmap and they will be 1.0 when they reach certain goals.

I like this comparison, because to me Inkscape has pretty much everything I need. I don't know why Inkscape didn't transition to 1.0 version earlier. I actually don't care. What I care is that it works and has worked for me for the past decade. This is what I mean about the quality. Inkscape's quality speaks for itself, so the version, being stuck at 0.48 or 0.92 is not so important as the features contained within the software. It would be nice that it had a 1.0 release to show its maturity but even that is secondary to its function.

One of my cousins is asking for a new version of Photoshop and whatever. Well, I could suggest her Inkscape, but what do I care? It's her money that she is wasting not mine.

Now, I'm not a professional illustrator. I've certainly read complaints about GIMP and Inkscape being insufficient for professional work. But again, what do I care? They work for my needs, and they are free, that's good for me.

If the topic of v1.0 keeps coming up perhaps it is time to have a roadmap? ...

Again, we don't have a structure in place (FreeCAD Organization). Who is going to make such roadmap, and execute it? Unless somebody organizes this for real, such roadmap is not going to happen. I have my own personal roadmap in the Draft/Arch side; it will take me around 5 years to execute it, unless I see more programmers joining the effort, like carlopav. Since Draft/Arch is basically the work of just a few people, such roadmap hasn't been necessary, we can basically commit anything to the workbench that "makes sense".