This was discussed some time ago as I remember.... As a matter of interest, who run kicadlib.org s that would seem to be an obvious starting place.
I don;t think an automated system will work 100% but it is the obvious route for most operations. You would probably end up with a two stage system. The "normal" system where things are uploaded by contrib. when can be anyone, then a second set which has been audited as ok and locked to prevent any changes to the list. (obviously what a user does after download is up to them) What you must not allow is parts to be uploaded under any name or description, that causes chaos, so whatever system you use must allocate the critical names and identifiers. I've seen the results of allowing people to do that, and it's not nice - people would fill in equipment records with things like Cisco Hub, cisco switch, cis. switch and so on. This made finding things very hard. That was soon changes to a drop down list so that everyone picked the same name for a particular equipment type. The hardest part is just defining the minimum requirements that you really need. A part is only one file, and a module is maybe two if you include a 3d file as well. Including an image of the part or mod might be something to think about but that's open to debate. You also need to define a set of criteria that defines when a part is deemed to be audited as OK. It will not be sufficient to say that x people have use it, and say OK, as an example from the group over the past few weeks, a member found a problem where the silk screen covered the pads. Not an obvious thing to think of, but important for pcb production house builds. What you really want to avoid is getting bogged down in details that make things overly complex. Andy On Tue, 02 Feb 2010 08:40:28 -0500 mike <m...@pikeaero.com> wrote: > Anders, > > Yes, I think that would be possible, however, I think you would wind up with > something > like the yahoo file area where you just have everything sort of scattered > about unless > you have a human gatekeeper of some sort. > > So I guess that's the first requirement, should it be automated or include a > human > gatekeeper of some sort? > > Personally, I would prefer an automated or semi-automated type of web based > solution. > > My vision of a KiCad Parts Repository: > > Uploading parts: > > I goto http://kicadparts.org or whatever and create an account. > I upload some widgets. > Before the widgets can be published, I need to classify them > into some sort of category structure. I need to provide > a jpeg image, and a brief description. > > Downloading Parts: > > I goto http://kicadparts.org or whatever and type into the search box "PLCC" > and I get a list of all the part hits that are PLCC sockets of some sort. > Maybe there's a refined search where I can narrow the search. > > The downloading parts should even be almost like a shopping cart experience... > I'll have one of those, one of those, I wonder if they have something like > this.... > > You should be able to select categories of parts and have them assembled > into a lib before downloading. Or how about even you add parts to a basket > just like a shopping cart, and then you have the option of assembling parts > in your basket into a lib at "checkout".... > > Feedback on parts: > > You need some method of QA on parts...I suppose in a community based > structure the best QA is some sort of feedback mechanism > where you can leave feedback on parts that others can see. > > I suppose the parts should be able to have some sort of revision history > so that that consumers can see if issues have been addressed with a > particular part, etc.. > > My point is that if people are going to consistently use the repository, > it has to be user friendly, custom tailored to this specific application...I > think. > > There are a few issues with what I have proposed like the handling of known > defective parts and non-responsive part owners, and things like that, but > those sort of things would probably just fall to whomever is the admin as > sort of policy issues I think. > > I guess what I am proposing, is a custom, one-off web based application > solution. I know, it's a lot of pain up front but I think it would really pay > off > in terms of boosting KiCAD to a new level. > > Comments? > > Mike Sharkey > > > On February 2, 2010 02:19:51 am you wrote: > > I think it is an excellent idea. Why not use sourceforge? > > > > - Anders Gustafsson > > Engineer, CNE6, ASE > > Pedago, The Aaland Islands (N60 E20) > > www.pedago.fi > > phone +358 18 12060 > > mobile +358 40506 7099 > > fax +358 18 14060 > > > > > > > > >>> mike <m...@pikeaero.com> 2010-02-01 22:37 >>> > > I would not mind lending a hand to coordinate some sort of repository > > effort if we can > > agree on what a KiCAD parts repository might look like for starters. > > > ------------------------------------ > > Please read the Kicad FAQ in the group files section before posting your > question. > Please post your bug reports here. They will be picked up by the creator of > Kicad. > Please visit http://www.kicadlib.org for details of how to contribute your > symbols/modules to the kicad library. > For building Kicad from source and other development questions visit the > kicad-devel group at http://groups.yahoo.com/group/kicad-develYahoo! Groups > Links > > >
