Community opinion polling central thread

Have some feature requests, feedback, cool stuff to share, or want to know where FreeCAD is going? This is the place.
Forum rules
Be nice to others! Read the FreeCAD code of conduct!
User avatar
uwestoehr
Veteran
Posts: 4961
Joined: Sun Jan 27, 2019 3:21 am
Location: Germany
Contact:

Re: Community opinion polling central thread

Post by uwestoehr »

openBrain wrote: Sun Oct 16, 2022 4:11 pm Over and out.
Please. We have some missing information. Therefore we wil meet now and sort it out, agree to something, make the decision public and then we can move on.

We both worked together now for years I think and now we have an issue for the first time. I am sure we can resolve this. Tensions are normal, and that's why it is important to meet regularly. That's why we have now a quarterly meeting where we look each other in the face and can debate as we like.
Therefore please join the meeting.
User avatar
sliptonic
Veteran
Posts: 3457
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Community opinion polling central thread

Post by sliptonic »

openBrain wrote: Wed Feb 09, 2022 5:40 pm This thread proposes to improve this by centralizing links to community "polls" inside the various threads of the forum.
This has been talked about in the past and I expressed some reservation about it then. I still think it's problematic for a couple reasons.

First, it's bypassing the CONTRIBUTING process both in method and intent. We're not clearly identifying a problem and then proposing a solution, rather we're asking for opinions and then acting based on votes.

Second, there's no rigor around the voting. I understand that this is just meant to get a sense of how the community feels about an issue and isn't binding. I don't think that's good enough. The FreeCAD community has grown well beyond the forum with large groups on linkedin, Reddit, Facebook etc. They have no visibility to the polls. More importantly the polling doesn't reflect any kind of demographics. Are developers well represented? What about commercial users? What about first time users vs experienced users?

In the end, it's impossible to look at a poll result and understand what it means.

I think it's great that we're trying to get some consensus around these subjective issues but I'm concerned that polls don't give useful information and might give a false sense of consensus.

IMHO there's still far too much development discussion happening in the forum where important points get lost or buried and far too little in the issues where they can make a difference. For example:

Here's an forum topic that has gotten 75 comments in two weeks. No corresponding issue in the tracker.
Here's a tracker issue that has gotten 20 comments in four months.
GeneFC
Veteran
Posts: 5373
Joined: Sat Mar 19, 2016 3:36 pm
Location: Punta Gorda, FL

Re: Community opinion polling central thread

Post by GeneFC »

sliptonic wrote: Tue Feb 14, 2023 5:37 pm IMHO there's still far too much development discussion happening in the forum where important points get lost or buried and far too little in the issues where they can make a difference.
I understand the utility of moving specific discussion to github.

The problem is that the github discussion interface is dreadfully awful. All sorts of automatic check reports and other non-discussion stuff interspersed into the discussion.

Probably no real way around it, but it makes following the conversation quite a bit more difficult than it should be.

Gene
openBrain
Veteran
Posts: 9034
Joined: Fri Nov 09, 2018 5:38 pm
Contact:

Re: Community opinion polling central thread

Post by openBrain »

sliptonic wrote: Tue Feb 14, 2023 5:37 pm
openBrain wrote: Wed Feb 09, 2022 5:40 pm This thread proposes to improve this by centralizing links to community "polls" inside the various threads of the forum.
This has been talked about in the past and I expressed some reservation about it then. I still think it's problematic for a couple reasons.
Sorry, can't get where you're going with all that. :?
First, it's bypassing the CONTRIBUTING process both in method and intent.
I don't think it is. But will come back on this.
This thread is by far prior to contributing process. Even the last post before you revived it is. So if you think contributing process has made this useless or obsolete, just redirect to the former and lock the thread. Over.

Anyway, if you read the OP, this doesn't aim at having any role in the contributing method, but only bring a tool that can facilitate achieving a choice in some (described) situations. It doesn't state poll results will give any power about anything. ;)
We're not clearly identifying a problem and then proposing a solution, rather we're asking for opinions and then acting based on votes.
Not clear about what you're talking.
* If you talk about this thread, it clearly identifies a problem and proposes a solution. This is mainly what the OP is all about.
* If you talk about polls, you're comparing 2 things that lies on 2 different levels. Poll is a tool, and it's perfectly feasible to clearly describe a problem in it, and offer choice between clearly defined solutions. This is matter of using the tool. Which is case for any tool BTW.
Second, there's no rigor around the voting. I understand that this is just meant to get a sense of how the community feels about an issue and isn't binding. I don't think that's good enough. The FreeCAD community has grown well beyond the forum with large groups on linkedin, Reddit, Facebook etc. They have no visibility to the polls. More importantly the polling doesn't reflect any kind of demographics.
Well, that's bringing nothing new but confusion. The problem isn't well described and you're proposing no solution. :?
Please, what solution would you propose in such a solution : "I'm a developer and to fix a bug in FreeCAD, I have to choose my implementation between 2 different behaviors. I'd like to know which one is the more appropriate to community usage. How should I proceed ?"
Are developers well represented? What about commercial users? What about first time users vs experienced users?
Hmmmm. Should developers be considered as special users when we talk about FreeCAD behavior ?
"Commercial users" ??? There are no commercial users of FreeCAD. This is as saying "commercial users of Linux". It doesn't exist. There are only commercial users of products/services based on FreeCAD(Linux).

At this step, I feel like you're having very unclear intentions as trying to achieve some omnipotence on the contribution process to FreeCAD.
This feeling is certainly reinforced by your mixing roles as both maintainer (core dev) and Ondsel owner, and I can't prevent to think you're thinking about the latter leveraging of a well tailored contribution process.
This is a slippy slope IMO. :?
In the end, it's impossible to look at a poll result and understand what it means.
Again it's a matter of how the tool was used. The tool itself can't really be blamed.
I think it's great that we're trying to get some consensus around these subjective issues but I'm concerned that polls don't give useful information and might give a false sense of consensus.

IMHO there's still far too much development discussion happening in the forum where important points get lost or buried and far too little in the issues where they can make a difference. For example:
I'm curious how the tracker issue is more "demographic representative" than the forum...
Why not close all the development-dedicated subforums and point to GH tracker ?
Here's an forum topic that has gotten 75 comments in two weeks. No corresponding issue in the tracker.
Here's a tracker issue that has gotten 20 comments in four months.
There is probably as much bias in this example as in a poll (if not more). Maybe the first topic is highly interesting and/or dividing, and the second is absolutely not (don't know actually, didn't clicked the links :P)

PS: As reviewing this post, it may sound a bit rude. But maybe it shall. Don't know.
And don't think I have special concerns about what's happening to this thread. I don't. I'm happy we tried this and I don't care it's future. ;)
User avatar
sliptonic
Veteran
Posts: 3457
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Community opinion polling central thread

Post by sliptonic »

GeneFC wrote: Tue Feb 14, 2023 7:51 pm
sliptonic wrote: Tue Feb 14, 2023 5:37 pm IMHO there's still far too much development discussion happening in the forum where important points get lost or buried and far too little in the issues where they can make a difference.
The problem is that the github discussion interface is dreadfully awful. All sorts of automatic check reports and other non-discussion stuff interspersed into the discussion.
True for PR discussions but not for issue discussions. Yes it sucks. Yes we need to improve it (as much as we can) but ignoring it doesn't really help.
openBrain wrote: Tue Feb 14, 2023 7:55 pm This thread is by far prior to contributing process. Even the last post before you revived it is. So if you think contributing process has made this useless or obsolete, just redirect to the former and lock the thread. Over.
It seems like very development-oriented discussion continues here for a long time and a relevant issue capturing all those details never gets created. Why should it? With all the information here in the forum, copying it to the issue is a waste of time.

Being heavy handed and locking a topic is possible but I doubt that's how we want to run things. Doesn't seem very friendly.
Anyway, if you read the OP, this doesn't aim at having any role in the contributing method, but only bring a tool that can facilitate achieving a choice in some (described) situations. It doesn't state poll results will give any power about anything. ;)
As I said before, I understand that the poll isn't binding. What I don't understand is how it 'facilitates achieving a choice' How do you propose to interpret the results?
* If you talk about this thread, it clearly identifies a problem and proposes a solution. This is mainly what the OP is all about.
If it clearly identifies a problem, why isn't it an issue? If it proposes a solution, why isn't it doing so in the form of a PR?
Please, what solution would you propose in such a solution : "I'm a developer and to fix a bug in FreeCAD, I have to choose my implementation between 2 different behaviors. I'd like to know which one is the more appropriate to community usage. How should I proceed ?"
By all means, ask for feedback. That's not the problem. But in three years when someone is revising the code, how will they find the discussion and rationale that led to the final choice? At least if it's in an issue, it's linked to a PR, which is linked to the code. I suspect that the quality of the feedback in the issue will also be higher because it won't be purely subjective. You'll get developers offering feedback that might have more to do with technical considerations, future-proofing, UI consistency, and upstream considerations.

The only advantage that the poll offers (that I can see) is that it's extremely easy to create.
Are developers well represented? What about commercial users? What about first time users vs experienced users?
Hmmmm. Should developers be considered as special users when we talk about FreeCAD behavior ?
"Commercial users" ??? There are no commercial users of FreeCAD. This is as saying "commercial users of Linux". It doesn't exist. There are only commercial users of products/services based on FreeCAD(Linux).
My point is that different people can have differing valid viewpoints. Duh. The poll doesn't tell you if ANY of those viewpoints is represented.
At this step, I feel like you're having very unclear intentions as trying to achieve some omnipotence on the contribution process to FreeCAD.
This feeling is certainly reinforced by your mixing roles as both maintainer (core dev) and Ondsel owner, and I can't prevent to think you're thinking about the latter leveraging of a well tailored contribution process.
This is a slippy slope IMO. :?
Not omnipotence. The process shouldn't favor anyone. It should be rigorous and transparent. Absolutely I want a well tailored and fair contribution process. I hate the fact that contributions sit for months without getting merged. I hate that FreeCAD development has been slowed by infighting because we can't agree on things. If having those viewpoints means I have a conflict of interest, please clarify.


There is probably as much bias in this example as in a poll (if not more). Maybe the first topic is highly interesting and/or dividing, and the second is absolutely not (don't know actually, didn't clicked the links :P)
That's a fair criticism. I may have cherry picked examples. But I just grabbed two that were both about subjective UI things. Very few discussions on the tracker progress beyond a few comments but forum discussions frequently get to 20 pages.

Honestly, it isn't the polls themselves that I care about. I genuinely want to find a process that leads to consensus and avoids conflict. I just don't see how the polls achieve that. The tracker might not work either.
User avatar
adrianinsaval
Veteran
Posts: 5541
Joined: Thu Apr 05, 2018 5:15 pm

Re: Community opinion polling central thread

Post by adrianinsaval »

GeneFC wrote: Tue Feb 14, 2023 7:51 pm I understand the utility of moving specific discussion to github.

The problem is that the github discussion interface is dreadfully awful. All sorts of automatic check reports and other non-discussion stuff interspersed into the discussion.

Probably no real way around it, but it makes following the conversation quite a bit more difficult than it should be.

Gene
Regarding this I'll come back to an old request of mine, @bernd should the bot comments from the Gitlab CI be disabled? IMO it only pollutes the discussions in PRs, in some cases excessively so.
What do other people think about this?
openBrain
Veteran
Posts: 9034
Joined: Fri Nov 09, 2018 5:38 pm
Contact:

Re: Community opinion polling central thread

Post by openBrain »

sliptonic wrote: Tue Feb 14, 2023 8:57 pm Being heavy handed and locking a topic is possible but I doubt that's how we want to run things. Doesn't seem very friendly.
That's not a question of friendliness. It's about taking appropriate decisions. It can be done in a friendly way.
Nothing worse than taking no decision.
As I said before, I understand that the poll isn't binding. What I don't understand is how it 'facilitates achieving a choice' How do you propose to interpret the results?
The proposed solution with the more votes wins !
* If you talk about this thread, it clearly identifies a problem and proposes a solution. This is mainly what the OP is all about.
If it clearly identifies a problem, why isn't it an issue? If it proposes a solution, why isn't it doing so in the form of a PR?
It's not in an issue because it was done before we made it the process to use.
A PR to what ? It isn't about code, it's about a way to help choose the community-accepted solution. :?
My point is that different people can have differing valid viewpoints. Duh. The poll doesn't tell you if ANY of those viewpoints is represented.
Feel free to propose another tool that do. ;)
Very few discussions on the tracker progress beyond a few comments but forum discussions frequently get to 20 pages.
And ? This would need decent analyze to draw a conclusion from that. But I'm not paid enough to jump into doing this. :lol:
Honestly, it isn't the polls themselves that I care about. I genuinely want to find a process that leads to consensus and avoids conflict. I just don't see how the polls achieve that. The tracker might not work either.
Polls are easy to create, easy to use/understand, easy to check. For sure they aren't an entire solution, and maybe an inappropriate one.

OK, honestly I don't really want to consume more time about this. And still I don't understand why you revived this topic just to criticize it, especially given that it clearly hasn't been used since contributing process has been set up.
IMO, time for another developer meeting.
openBrain
Veteran
Posts: 9034
Joined: Fri Nov 09, 2018 5:38 pm
Contact:

Re: Community opinion polling central thread

Post by openBrain »

adrianinsaval wrote: Wed Feb 15, 2023 12:18 pm Regarding this I'll come back to an old request of mine, @bernd should the bot comments from the Gitlab CI be disabled? IMO it only pollutes the discussions in PRs, in some cases excessively so.
What do other people think about this?
Agree
User avatar
sliptonic
Veteran
Posts: 3457
Joined: Tue Oct 25, 2011 10:46 pm
Location: Columbia, Missouri
Contact:

Re: Community opinion polling central thread

Post by sliptonic »

openBrain wrote: Wed Feb 15, 2023 1:27 pm The proposed solution with the more votes wins !
Polls are easy to create, easy to use/understand, easy to check. For sure they aren't an entire solution, and maybe an inappropriate one.
And that's the only reason that I brought it up (again). There seems to be a sentiment that these things are harmless. But harmless is really the best case scenario. In fact, they may be counter-productive. There's actually some evidence of this.

This is a really great book. There's quite a bit of discussion about the impulse to vote and how harmful it can be in group conversations. Among other things, early votes tend to sway the outcome far more than is appropriate.
IMO, time for another developer meeting.
Absolutely!

openBrain wrote: Wed Feb 15, 2023 1:28 pm
adrianinsaval wrote: Wed Feb 15, 2023 12:18 pm Regarding this I'll come back to an old request of mine, @bernd should the bot comments from the Gitlab CI be disabled? IMO it only pollutes the discussions in PRs, in some cases excessively so.
What do other people think about this?
Agree
Agree
User avatar
pathfinder
Posts: 315
Joined: Mon Oct 04, 2021 10:27 am

Re: Community opinion polling central thread

Post by pathfinder »

Another point against polls:

They only represent active users, but not people who aren't users yet.

This may sound somewhat unintuitive, but I'm mainly talking about two groups:

1.: Future users

2.: People for whom FreeCAD isn't the right solution yet but has the potential to be in the future.

If you base your decisionmaking on already established users, established but unintuitive practices may be preferred over changing to more streamlined solutions that need to be relearned by established users.

As for the second group; an example I saw the other day was in the discussion around extrusion handles, direct model manipulation and so on, with one argument against it being that it counter to FreeCADs core principles to use direct model manipulation. While parametric modelling is the core utility of FreeCAD, other use cases and new user groups emerge for CAD. For example, Plasticity is a relatively new 3d modelling software that uses traditional CAD modelling with strong emphasis on direct manipulation to design game assets, such as guns, cars and so on and does not have a production focus.

I'm not saying that FreeCAD needs to cater to these new use cases specifically, but if it wants to broaden its user base, it may need to account for new and shifting demands that potential users have. If their demands are not met, they'll go elsewhere and not even contribute to the debate. You'll never know what these groups actually look for.

Now, this is my subjective perspective, both as someone who probably uses FreeCAD in a less "traditional" way than most users and as someone who believes that more users is, by and large, desirable. Both of these points are obviously debatable.
Post Reply