Low Cost PCB's Low Cost Components

Author Topic: [RANT] Why is it so hard to create a library?  (Read 1783 times)

0 Members and 1 Guest are viewing this topic.

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
[RANT] Why is it so hard to create a library?
« on: July 04, 2017, 10:37:49 PM »
I need to create a new part and it doesn't fit within the premade libraries. So, I naively look for the "Create library" menu option. Nope, doesn't exist. KiCAD expects you to update the libraries they provide. But here's the thing... Create a new text file, change the extension to ".lib", open it in KiCAD and viola! :palm: Why can I not do this from KiCAD? :rant:

Now to see if the same thing works for footprints. |O *SIGH!*
 

Online alm

  • Super Contributor
  • ***
  • Posts: 1075
  • Country: 00
Re: [RANT] Why is it so hard to create a library?
« Reply #1 on: July 04, 2017, 11:03:36 PM »
What is wrong with the "Save current component to new library" button?
 
The following users thanked this post: MarkS

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #2 on: July 04, 2017, 11:23:47 PM »
 :palm:

I need to sleep!
 

Offline SimonR

  • Regular Contributor
  • *
  • Posts: 96
  • Country: gb
Re: [RANT] Why is it so hard to create a library?
« Reply #3 on: July 07, 2017, 06:44:43 AM »
Don't feel bad I would have done the same thing.
Every CAD system I've used has a button for creating a "New Library" and I've used quite a few.
Combining it with something else is just confusing, why would you think to look for such a thing.
 

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #4 on: July 07, 2017, 08:23:54 AM »
This just illustrates my biggest gripe with KiCAD. The GUI is not intuitive. It doesn't follow industry standard interface design and for no good reason. This makes the learning curve unnecessarily complex. I seriously get tension headaches trying to use it. It just doesn't do what I expect when I expect it.
 

Online alm

  • Super Contributor
  • ***
  • Posts: 1075
  • Country: 00
Re: [RANT] Why is it so hard to create a library?
« Reply #5 on: July 07, 2017, 08:43:37 AM »
What is the industry-standard for CAD user interfaces? It seems to me like every CAD program has to have its own non-standard interface and keyboard shortcuts. For most users industry-standard just appears to be the CAD program they have spent the most time with.
 
The following users thanked this post: kripton2035, janoc, Jacon

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #6 on: July 07, 2017, 08:57:35 AM »
What is the industry-standard for CAD user interfaces? It seems to me like every CAD program has to have its own non-standard interface and keyboard shortcuts. For most users industry-standard just appears to be the CAD program they have spent the most time with.

Fair enough.
 

Offline tycz

  • Contributor
  • Posts: 49
Re: [RANT] Why is it so hard to create a library?
« Reply #7 on: July 07, 2017, 04:11:22 PM »
What is wrong with the "Save current component to new library" button?

The feature is not in the File menu where one would expect it to be.
 

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #8 on: July 07, 2017, 05:17:54 PM »
What is wrong with the "Save current component to new library" button?
The feature is not in the File menu where one would expect it to be.

This.

I guess instead of "industry standard", I should have said "GUI standard". User interface standards are well defined and have been for decades. KiCAD seems to be forging ahead as though those standards do not exist. Instead of using the fact that no CAD software follows any standard as an excuse to continue the trend, maybe it is time for CAD software to adopt the standards that the rest of modern day software follow.
« Last Edit: July 07, 2017, 05:19:43 PM by MarkS »
 
The following users thanked this post: garnix

Online alm

  • Super Contributor
  • ***
  • Posts: 1075
  • Country: 00
Re: [RANT] Why is it so hard to create a library?
« Reply #9 on: July 07, 2017, 11:13:40 PM »
No argument here. However, ranting on a forum where no / few KiCad developers are active is not going to change that. See here for how to get involved.
 

Offline MarkS

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #10 on: July 08, 2017, 12:06:10 AM »
No argument here. However, ranting on a forum where no / few KiCad developers are active is not going to change that. See here for how to get involved.

True, but their standard reply usually goes along the lines of, "That's the way it is and it is going to stay," or "If you don't like it, contribute." The first is patently dismissive and the second assumes decent programming skills of the person making the criticism. Both are said merely to shut down the conversation. Believe me, I've tried and gave that up. All I was doing here was venting and trying to get the tension headache to ease.
 

Offline Gibson486

  • Regular Contributor
  • *
  • Posts: 65
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #11 on: July 12, 2017, 10:44:34 PM »
Really? I actually like KiCAD. We moved onto Altium. There have been a few occasions where we went, "KiCAD does this better". My only complaint was that the GUI did feel Apple like, but once you get over it, it is fine.
« Last Edit: July 12, 2017, 10:48:00 PM by Gibson486 »
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1768
  • Country: fr
Re: [RANT] Why is it so hard to create a library?
« Reply #12 on: July 13, 2017, 01:03:14 AM »
True, but their standard reply usually goes along the lines of, "That's the way it is and it is going to stay," or "If you don't like it, contribute." The first is patently dismissive and the second assumes decent programming skills of the person making the criticism. Both are said merely to shut down the conversation. Believe me, I've tried and gave that up. All I was doing here was venting and trying to get the tension headache to ease.

I understand your point, but you also need to keep in mind that the Kicad team is rather small and unpaid (emphasis mine). They are doing this as their pet project/hobby/whatever for the most part. In that situation having people constantly barge in the mailing list/forum/IRC channel complaining about this or that in the UI and how software XYZ does it better, demanding fixes and how they must change their ways or nobody will keep using their tool gets old fast.

You may feel offended, feeling that your input is not taken as valuable and see it as an attempt to shut down the conversation (because it is exactly that). However, try to develop a moderately successful open source project that is used by a lot of people and you will see how long you will keep the development communication channels open to the "outsiders". All that valuable (for you) input is mostly a huge distraction. Sometimes the project leaders/developers need to put their foot down and be a bit rude. It is the only way to survive.

We may not be used to it, because we are used to customer support and companies bending over backwards for their clients and being justifiably outraged if they do not after we have parted with our money. However companies have armies of paid support drones to take the abuse and shield the people doing the actual work from all this. OSS projects usually don't. And with few exceptions we usually aren't their paying customers neither!

These projects rarely have the resources to deal with such requests, so the demand to contribute code is very reasonable. It is pretty much the only way to get things fixed if the core team has other priorities (like adding the missing functionality or fixing data destroying bugs). And setting those priorities is their prerogative too - it is their project.

If you aren't programmer and cannot contribute in that way, there are other methods - e.g. doing a proper usability study and suggesting a comprehensive overhaul of the UI/workflow. However, that needs to be founded on something more than just "I prefer having this button there and called that." Or you can do bug triage. Or simply donate to the project. Any of this is better than venting on a public forum about a tool you are using for free.

« Last Edit: July 13, 2017, 01:10:44 AM by janoc »
 
The following users thanked this post: jakeisprobably

Online 2N3055

  • Frequent Contributor
  • **
  • Posts: 707
  • Country: hr
Re: [RANT] Why is it so hard to create a library?
« Reply #13 on: July 13, 2017, 04:31:30 AM »
Janoc,

first I want to say I have utmost respect for people that donate anything for the betterment of the humanity.. Time, money, software skills, whatever.

But most charity organisations will actually not accept just anything that you think might be useful. They will accept only money, non perishable goods and important stuff. Because they (and people they help) have no need for Gucci bags or such. They need other things...

Basically, what you're saying is, Open Source programmers are doing something for their own reasons, priorities, and whatever they do we have to accept because we didn't pay for it. It was, like charity, a gift, and we do not get to criticize what was so graciously given to us, with great effort and completely free...

And Open Source programmers don't care what poor users that cannot afford expensive software say... Users are not to complain.. It is free, and even if it is almost unusable except for proof of concept, Open Source programmers will not try to make it real software that users can actually use...  Nooo, because, they are not really interested if anybody will ever use it, they are just writing it to show how smart they are, and for the bragging rights..

I must have misunderstood you (my comprehension of English is limited, sorry), because last time I checked, Open Source was meant to make viable replacement for overly expensive commercial products. so everyone, even the poorest one, can have the same chance for education, better life and such...

I use Open Office and find it easier to use that new MS Office with retarded Ribbon... LOL Teletubbies...

Kicad I tried, and couldn't use it. It is pretty much unusable in it's current form. It is technically capable, but usability is not there.
It kinda works, but you cannot achieve productivity in it...
It is currently in a stage of a proof of concept software.
Team proved they are fantastic programmers, capable of writing fantastic, technically sophisticated software. They are. Kudos.
But Kicad has no workflow, user interface is confusing (that's being kind), Libraries and their management is just.....

Great software is not just about clever code and algorithms.. Great software is one that is great to use, intuitive, quick, useful....

So no, I'm not ungrateful to a gift given to me.. I politely refused to take it, because in it's current state it is useful to me as ice on Antarctica , and sand in Sahara..
And  if it is helpful to someone I'm happy for them.
To me, it isn't, and won't be until it is made useful in addition to free..

In which case I would be glad to financially support the effort, to help develop it, and to help it be available to those that are not fortunate like me....
And no, I cannot contribute in other way, my time is limited, and have no programer skills on the level to be worthy to join the team..

I'm not prepared to donate to the project that I think goes nowhere with it's current direction,  that is not very useful to me, or most people out there..

Maybe, in the future, Kicad team will be able to make it more complete and such and I will change my mind ...
I'm not saying I can understand their vision, and maybe they are going in great direction and Kicad is destined for greatness.. I hope for that.. I really do.


But for now i don't see it.

Regardless, to developers, thank you for all of your selfless and hard work, but no thank you. Maybe some other time.

Best regards,

Sinisa
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 1768
  • Country: fr
Re: [RANT] Why is it so hard to create a library?
« Reply #14 on: July 13, 2017, 08:00:46 AM »
But most charity organisations will actually not accept just anything that you think might be useful. They will accept only money, non perishable goods and important stuff. Because they (and people they help) have no need for Gucci bags or such. They need other things...

I believe I didn't speak about donation of Gucci bags but of code contributions, monetary contributions or even efforts to improve the project. I don't know how Kicad but e.g. Mageia Linux (Linux distribution) often calls for volunteers to help with bug triage, testing, general QA, etc.

Kicad needs people to help with documentation and even with component library maintenance:
http://kicad-pcb.org/contribute/docs-team/
http://kicad-pcb.org/contribute/librarians/

And Open Source programmers don't care what poor users that cannot afford expensive software say... Users are not to complain.. It is free, and even if it is almost unusable except for proof of concept, Open Source programmers will not try to make it real software that users can actually use...  Nooo, because, they are not really interested if anybody will ever use it, they are just writing it to show how smart they are, and for the bragging rights..

Most open source projects started because the creator needed to scratch their own itch. Of course they do care that people use it but it is not a product. The developer can stop working on it tomorrow and guess what - he or she doesn't owe you anything. But feel free to take the code and continue working on it  (or hire someone else to do it for you) - that's the liberty that free (as in freedom) software guarantees you. They are not making the software for you but first and foremost for themselves. The program being useful for someone else is in most cases a bonus.


I must have misunderstood you (my comprehension of English is limited, sorry), because last time I checked, Open Source was meant to make viable replacement for overly expensive commercial products. so everyone, even the poorest one, can have the same chance for education, better life and such...

That you have really misunderstood because that has never been a goal. Open source/Free software is not about replacing expensive commercial software with free (as in beer) programs so that you don't need to buy Windows or Word. It has always been about the freedom to modify the code and to tinker with it, so that you aren't screwed when the product doesn't do what you want or when the manufacturer stops supporting it or the manufacturer explicitly doesn't want you to tinker with it - such as the various DRM schemes to milk money from you.

The fact that you don't have to pay for a lot of that software (open source != gratis, there is paid open source software too) and that it allows access to education and better life even to people who wouldn't be able to afford it otherwise is only a (nice) side effect.

Kicad I tried, and couldn't use it. It is pretty much unusable in it's current form. It is technically capable, but usability is not there.
It kinda works, but you cannot achieve productivity in it...
It is currently in a stage of a proof of concept software.
Team proved they are fantastic programmers, capable of writing fantastic, technically sophisticated software. They are. Kudos.
But Kicad has no workflow, user interface is confusing (that's being kind), Libraries and their management is just.....

I believe a lot of people would disagree with that statement - plenty of products and hardware has been designed with it. That doesn't mean that things cannot be improved but let's not to turn this debate into another "Why Kicad is shit"/"Why Kicad works, you just don't know how to use it" thread. That would be just  :horse:

So no, I'm not ungrateful to a gift given to me.. I politely refused to take it, because in it's current state it is useful to me as ice on Antarctica , and sand in Sahara..
And  if it is helpful to someone I'm happy for them.
To me, it isn't, and won't be until it is made useful in addition to free..

More power to you - there are plenty of other solutions out there. Fortunately, nobody is forced to use Kicad. Unfortunately in most cases you will have to pay for it. What you are describing as requirements/defects sadly costs engineering and development time to fix and that isn't free. If it happens to align with a view of one of the developers and they will do it, great. But you have no right to demand it or expect it unless you are paying for it or contributing to the project in some other way. That's not how things work.

Other vendors are asking 3-4 figures/seat for comparable software. Go figure why that is so ... Oh and good luck convincing a commercial software vendor to modify something to suit your workflow or something else without paying through the nose for the privilege.

In which case I would be glad to financially support the effort, to help develop it, and to help it be available to those that are not fortunate like me....
And no, I cannot contribute in other way, my time is limited, and have no programer skills on the level to be worthy to join the team..

I'm not prepared to donate to the project that I think goes nowhere with it's current direction,  that is not very useful to me, or most people out there..

Fine, but then don't expect them to change direction to suit your needs. If you, by your own admission, aren't willing to contribute to the project in any way, even though they are looking for help even from non-programmers then why should they fix things to fit your needs? For free, to boot. Bashing the project with how it is crap and unusable is not really in any way constructive in that case.
« Last Edit: July 13, 2017, 08:15:06 AM by janoc »
 

Online 2N3055

  • Frequent Contributor
  • **
  • Posts: 707
  • Country: hr
Re: [RANT] Why is it so hard to create a library?
« Reply #15 on: July 13, 2017, 08:53:52 AM »
Janoc,

Thank you for taking time to post a reply to what I wrote..
I obviously misunderstood the whole thing...
I do better understand some things now, and I thank you for explaining it to me..

I wasn't trying to bash Kicad. I was trying to say that it is promising, but not ready to be "used" as a real product.

And you just explained to me why is that, and why it never might be, that not being important objective of the project...
It will be developed at it's own pace and priorities, by people that carry project. Their work, their prerogative.
And that is only fair.

If, by chance, software they make suits your need, you are free to use it.
You are also free to join the effort.

Those who don't like these options, are free to buy something else or make their own.

Fair enough.

For me, this means I can't use it now ( I tried, I couldn't get used to it), and that even if I donate or join the effort, there is no saying whether it will be "useful" to me in a timeframe that makes sense to me... So I made good decision to buy something now ('cause I need it now, not in 3 years...)

And Kicad might get closer to what I need in the future, and then we'll see...

Thanks and take care,

Sinisa

 

Offline Bassman59

  • Frequent Contributor
  • **
  • Posts: 475
  • Country: us
  • Yes, I do this for a living
Re: [RANT] Why is it so hard to create a library?
« Reply #16 on: July 13, 2017, 10:00:59 AM »
I wasn't trying to bash Kicad. I was trying to say that it is promising, but not ready to be "used" as a real product.

A lot of people will disagree. A lot of people will also say that EAGLE is unsuitable for real work because of its daft user interface.

Quote
And you just explained to me why is that, and why it never might be, that not being important objective of the project...
It will be developed at it's own pace and priorities, by people that carry project. Their work, their prerogative.
And that is only fair.

For me, this means I can't use it now ( I tried, I couldn't get used to it), and that even if I donate or join the effort, there is no saying whether it will be "useful" to me in a timeframe that makes sense to me... So I made good decision to buy something now ('cause I need it now, not in 3 years...)

I see a lot of people like janoc quoting the standard open-source software party line about how the developers do what is interesting to them and all of that. And yes, that's true.

But ... I follow the Kicad developers' mailing list closely, and I do as much as I can to test new builds with new features. And yes, ever since the 4.0 stable was released, there have been many features added and many improvements -- a lot of which were suggested by users who were able to make a clear case as to why their pet feature was necessary. A good example is the addition of via stitching. Turns out the problem was more difficult to solve than users thought it would be, but several developers worked on it, proposed solutions, offered branches to be tested, and ultimately it was merged into the main product trunk. (NB those new features won't be backported to the 4.0 stable branch.)

And yes, the library system has its many detractors, as does the difference between the layout program's functionality and the schematic program's. There has been a lot of work on the schematic side, including making it work with OpenGL, ensuring that hotkeys are the same across both programs, and a revamp of the schematic library system (to have a library table like pcbnew). Even standard ctrl-V/ctrl-C copy-and-paste, a common complaint from users, is being implemented. Users also complain about the odd ways in which schematic parts are copied, and how there's no rational way to copy a symbol or group of symbols from one sheet to another, and those complaints are being addressed. Advanced 3D rendering using STEP models was something pushed by users and implemented, and works really quite well.

Anyway, as to developers outright rejecting user feature demands ... janoc has it mostly right, but what he didn't mention is this: the developers really have heard all of the feature requests before. They know what Altium offers, what EAGLE offers, what Diptrace offers. They really don't live in a bubble, ignoring all outside discussion. What they would hope, I think, is that before some new person barges in on the mailing list or a forum with feature demands, the new person should spend the time looking through the bug reports and wishlist and the developer mailing list to see if the feature was already requested, and perhaps to learn why it wasn't implemented. Maybe that must-have feature can't be implemented within the existing program architecture. Or maybe they just don't have anyone who has the time to implement the feature.



Professionals who've used other PCB CAD packages tend to come to Kicad and say, "what is this, having to associate footprints with a symbol after the schematic is captured! What is this 'CvPCB,' anyway?" And yeah, the tutorials targeted at makers and hobbyists seem to all use that design flow, and that makes it seem like it's the only way to go. But if you follow the user forum (https://forum.kicad.info) you'll see that a LOT of the users really don't like that flow, and instead will create library symbols with a callout to the footprint included. And as the library effort progresses, they are adding such components, all because users want them. So if you're put off by the CvPCB flow, you don't have to use it. Create your own libraries.
 
The following users thanked this post: janoc

Offline julian1

  • Frequent Contributor
  • **
  • Posts: 270
  • Country: au
  • http://blog.julian1.io
Re: [RANT] Why is it so hard to create a library?
« Reply #17 on: July 13, 2017, 11:55:55 AM »
My experience has been that using the Kicad GUI was difficult and unintuitive - and I had to spend time to learn actual workflow sequences. But the ability to drill-down and create/edit libraries and footprints in Vim or a basic text editor as well as track project changes in source-control made up for it.

Offline Gibson486

  • Regular Contributor
  • *
  • Posts: 65
  • Country: us
Re: [RANT] Why is it so hard to create a library?
« Reply #18 on: July 13, 2017, 10:35:26 PM »
Honestly, the complaints about the UI, I think are due to your familiarity with a different package. Once I got used to KiCAD's UI, it was not bad at all (does feel mac like, but whatever). Yes, there could be better integration between schematic and PCB, but it is not that much worse than what the paid software does. When I moved from KiCad to Altium, the UI for Altium felt so weird. Till this day, I still think library navigation in Altium is not really that much better than KiCad. Also, if you do schematic hierarchies, you see an instance where KiCad is better.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf