How to make Eeschema library editor much better - Wishlist

Take a look at this image:

As I spend a lot of time at the library developing, I’m dreaming of such additional side panel that would facilitate symbols selection for editing.
The existing pin table should also be reworked as a side panel or integrated with above panel as another tab.

ps. I suppose the CERN guys will love it, because it reminds that from DXP. :sunglasses:

1 Like

Hi @keruseykaryu
why not create a bug in launchpad?
https://bugs.launchpad.net/kicad/+filebug
and mark it as a wish.

… or you can try to develop a patch and propose it to be merged in kicad.

Because I know the devs answer.

I’m not a C++ programmer. At least at that level to supply a patch.

What does it mean?
So why post it here?

If you don’t ask for it you have no chance of getting it :wink:
It’s actually more of a blueprint that a wishlist bug:

I have notifications from bug tracker. I saw many conversations there and on the devs mailing list. Belive me, user wishes are not quite welcome if they not also contain the patches.
This problem does not concern only to KiCad, but also a large part of the Open Source community. And this is sad.

Because I want to know the opinion from the other users. Was this forum not the appropriate place to do this?

I miss this feature too. Coming from an Altium and Diptrace background.

Sad?
If you were a dev and you have the choice to spend your free-time on something that interests you or on something that somebody else needs and wants (for free), what would your priorities be?
Do you also go around in your neighborhood and clean public places, repair benches, remove dog poo, create a children’s playground?
Now imagine - while you’re doing that - some neighbor comes by and tells you that he imagines you doing it his way?
How would you find that?

[quote=“keruseykaryu, post:3, topic:2388”][quote=“kammutierspule, post:2, topic:2388”]
or you can try to develop a patch and propose it to be merged in kicad.[/quote]
I’m not a C++ programmer. At least at that level to supply a patch.[/quote]
That’s no excuse.

Imagine that neighbor, starting to help you fixing the benches and building that children’s playground… how would you think about his ideas how to do those things then?

Such an approach calls into question the idea of the creation of Open Source software for public use.

Is there any admin to remove this topic? The discussion goes to the wrong direction,

Ok, on topic…

I find the idea to have a permanent list of the symbols/footprints contained in a library/repository in the respective editors a wastage of screen real estate (and I already use 2 screens).
If one needs an overview like that he can use the browsers, that’s what those are there for.

That feature with the pin table looks interesting, but you didn’t put a pic up for that.
Would you be able to create some script in python that works as a proof-of-concept for that one?

PS: on OSS

The best way to implement this sort of thing would be with a dockable window framework. Then users can decide what panes they want and where. It is an effort to put in the framework, but once done is really powerful. I did it for an application I support at work, using an FOSS component (DockPanelSuite), and now I don’t have to worry about users asking for special displays, I can just tell them to customise how they want.

For this particular case, usually I spend most time editing a symbol, but recently I have been checking whole libraries, so a quick way to scan through all the symbols is really useful. I have written an external tool which can do this, it means I can also do some automatic checking and conversion as well.

+1 for dockable windows!
I believe they are on Eeschema roadmap.

This is a good discussion.
It would be good for some users that have ideas and suggestions to better understand the development process.
The fact that someone have a very good idea or that there are 1K users approve the idea, the steps to make it happen could take very long and lots of efforts.

Exactly, what do you think it is better? A .bmp mockup to be discussed or a really working patch to be tested and integrated in the main trunk?

How do you propose that this could be handled? How long should it take? How many developers should be (by their own initiative) to be allocated to work on users wishes?

I become a KiCad developer, because no one will implement my wishes. So I learn that if I want it, I have to do it.
The funny thing is that the fact that it is an open-source and should be something goodwill, it is in fact a very selfish activity, because people only take efforts on their own needs and interests.

So I only see two ways of get wishes to reality: Or you do it yourself, or you could move (the interest of) some developer to do it for you (that can be for free or a paid activity)

Cheers!

Not for the librarians (developers, maintainers). Clicking every time to call a separate dialog with the list or clicking once more to call a browser can be annoying.

The pin table viewer (I think it could be extended to easy edit pins) is implemented in recent KiCad releases.

For expression your own ideas: no matter how.

I’m probably going to be sorry I dove into this mess but here goes.
I am both a user and author of Open Source.

I have seen this “friction” occur over quality, features, stability, usability, etc, many times.

What I see happens is a new user to a product (like myself) may not have known the development history, and the developers only hear the complaints and feel unappreciated.

I wouldn’t be surprised if 3-4 years ago KiCad users and developers where pretty happy to have the damn thing work at all, and any kludge was better than nothing. As time goes on the product gets better and some of the old band aids that most were happy to have in the first place, start being viewed derogatory. As a product matures and becomes more complicated the design cycle slows and users fall into the trap of expecting commercial quality service.

Over the years I have primarily used CircuitMaker, Autotrax EDA, DesignSpark, and Fritzing. I tried Kicad, 2 years ago and passed on it, I tried it again a year ago and passed on it, this year it meets my expectations and I stopped paying for Autotrax EDA. If we (the users) forget how much we pay for the development it is counter productive. If something does not meet our needs, we should move on and go to something else (come back in a year); be happy for those that can use it.

I am a "ASM, C, and PHP programmer. I never did like OOP, but I know the effort and complexity needed to add dockable windows to an existing project. Ya, I agree I get frustrated with extra clicks (I think the home grown editor on this forum sucks. But the guy that programmed it is probably dancing in circles to get it where it is.) but it is far better to be part of the solution than demoralizing the work crew. As a new user I have noticed quit a bit of inconvenient things but it was worth not paying the other guys.

The neatest component library file system I have seen is DesignSpark, but I am happy the KICAD developers took more time making things stable than adding features that break.

I’m a stranger here, and it is none of my business but I think it would make sense to prioritize a wish list so the users understand their input is heard. (They will often need to be reminded that there is a difference between wanting and getting), but the developers need to understand that users have no concept of the effort to fix what appears to be a small thing.

[TRUE STORY: In the 1980’s I used Borland Pascal and they had a steep development curve. Each release users complained about several nagging features. There was one feature that was really irritating and the dev team had told the CEO (Phillip Kahn) they estimated it would take 3-4 programmers 6mo to fix - other features and fixes deserved the resources. Two years later the CEO ask the development team to take another look and give a very accurate estimate. The dev team dispatched a programmer to look at the problem and three days later he accidentally stumbled over some code and fixed the whole thing. My experience was always I estimated 10 days and it ended up 6 weeks. You just don’t know what lurks under the hood]

The moral of the story was everybody at Borland had this feature on the top of the list but it appeared to take too much resources. I would rather have 5 stability issues fixed than receive the dockable windows - but the next guy will see things different. I am pretty sure I have lost 6 years of my life; 3 seconds at a time using unneeded mouse clicks - Oh! the humanity! - THANK YOU FOR ALL THE HARD WORK KICAD TEAM :slight_smile:

5 Likes

Off topic, but if KiCad used github, then they could use BountySource. While it is true most developers only work on their pet interests, there are some “freelance” developers who may be looking to learn a new project or exercise their skills, and are looking for random bug fixes/features to work on. For a few bonus $$, it’s enough to give an incentive for some devs. It also means users have a way of financing developments without directly hiring a programmer. There may well be people for whom $200 is a decent amount of money where they are.

BountySource take a percentage, so you might prefer to donate direct to Kicad, but then you have little input as to priorities. Also, there needs to be cooperation between the “bounty hunter” and core developers to ensure patches are acceptable.

I think that for the Kicad libraries that are hosted on github, BountySource could be used now. e.g. https://www.bountysource.com/teams/kicad/issues?tracker_ids=638943

This is a great idea - it makes me wonder why all open source software doesn’t offer this. I am a db programmer and a work list is trivial. If paypal was used, all it costs is 3% no one would forfeit a fee. There would be no credit card processing or security liability. The only thing is an arbitrator would have to be agreed on for disputes.

I have used freelancer.com for areas I am not skilled in (like PHP template engines) and they generally used 50% down and the rest upon completion.

Yes.

But make it parts and not symbols.

Symbols reside under parts.
As do footprints.

Parts tie symbols and footprints together into a… Part.

It seems that the Library Editor will be updated and incomming changes go in the right direction.

Dev mailing list thread: https://lists.launchpad.net/kicad-developers/msg26896.html

5 Likes

Video looks nice, (no sound?) and lists these changes