Speak EV - Electric Car Forums banner
41 - 60 of 70 Posts

·
Driving yet another EV!
Joined
·
2,578 Posts
2. http://openchargemap.org/site/develop#api seems clear enough to me. What's the issue?
This line is the crux:
Data imported from 3rd party Data Providers is copyright the original Data Provider in each case and is not provided under the same terms as the user-contributed data detailed above.
Any developer would be mad to use OCM data while this clause exists. It means that using OCM data requires you to either filter out 3rd party data (and thus deliberately have an incomplete map) or constantly check which 3rd party providers have added data to OCM and then go and negotiate a licence with each and every one of those providers.
 

·
Registered
Joined
·
1,904 Posts
I admit I don't like the third party data.
IMHO it's of poor quality anyway but I think it's there just to get enough info to make the map useful.
I'd like to get a to point where all that data can be removed from the system as it has been replaced by proper crowd sourced data.
The main source of the third party data seems to be the NCR is as we all know is of very low quality.
 

·
Registered
MG ZS Exclusive
Joined
·
238 Posts
By quality I mean both verified location, but also really useful stuff. Like exactly which RFID card I need to operate it, what the parking charges actually are, probability of ICEing, last known good use date, a real photo (not a google street view of a dumpster) etc. (PlugShare is best at this...but I'm sure can do better.)

I am doing exactly the above for my local area, updating OCM. Do I continue to do so? Am I wasting my time?
 

·
Registered
Joined
·
647 Posts
Regarding the GPL vs BSD question you'll find me in the BSD camp because the selfishness of others is nothing to be feared. If they can take my contribution and make something better and profit thereby and don't want to share-back, then that's their prerogative. Ultimately the contest between "open" and "proprietary" is a race with regards to users. Licensing is relevant only as it impacts usability and user adoption. Whether you're "open" or "proprietary" becomes largely esoteric once you have enough users.
 

·
Registered
Joined
·
1,904 Posts
By quality I mean both verified location, but also really useful stuff. Like exactly which RFID card I need to operate it, what the parking charges actually are, probability of ICEing, last known good use date, a real photo (not a google street view of a dumpster) etc. (PlugShare is best at this...but I'm sure can do better.)

I am doing exactly the above for my local area, updating OCM. Do I continue to do so? Am I wasting my time?
Yes and absolutely not.
If you need to, create a new point even if it's near to the incorrect one. We're having discussing with the OCM database owners to work out a way of linking the user entered data and being able to link/retiring the imported data.
I will certainly trying to do my bit for OCM and have done a few local points. I will be doing more.
The ultimate goal is to eliminate any automatically imported data which fixes the third party license issues as a side effect. This is going to require a fair sized crowd effort. The submission for OCM needs improving and streamlining to help with this.
 

·
Registered
MG ZS Exclusive
Joined
·
238 Posts
Yes and absolutely not.
If you need to, create a new point even if it's near to the incorrect one. We're having discussing with the OCM database owners to work out a way of linking the user entered data and being able to link/retiring the imported data.
I will certainly trying to do my bit for OCM and have done a few local points. I will be doing more.
The ultimate goal is to eliminate any automatically imported data which fixes the third party license issues as a side effect. This is going to require a fair sized crowd effort. The submission for OCM needs improving and streamlining to help with this.
I dont understand much of what you said BUT so long as the data I enter is not deleted AND is useful then I will continue to enter it
 

·
Registered
Joined
·
62 Posts
Hi Guys,

Very interesting discussion! I published our OCM data feed as KML to Map Gallery because Google asked if I wanted to. I did it pretty quickly and more of as an experiment - it is currently refreshed when I push the update button on the map draft so more recent content from OCM will be updated on that map each time. I'm in discussion with google on how best to automate that process. This is not 'another' charge map - if it says 'Open Charge Map' then your looking at something that can/will be kept reasonably up to date.

Presentation-wise, this is the raw KML output from our API and as such hasn't had work on presentation other than cleaning up the description formatting a bit - logically we'd use icons/colours in some way. I haven't yet had time to explore this.

If you see data you think is rubbish please follow the link to the details and update them - everyone can, just Sign In to edit. This edit goes into an approval queue and if it looks reasonable we approve it, then the various sources that use OCM (including) this eventually get the updated data as and when they request it.

The OCM app (as discussed elsewhere) is always a work in progress and our next version on mobile devices is roughly equivalent to http://openchargemap.org/app/ and will be using native google maps on Android and iOS. We are in the process of working out refinements to the user interface to ensure we have parity with other popular apps, but we can't really change your brand preference if you have one, that's up to you.

I don't mind if people use PlugShare, thats OK - your data just goes into a silo that no-one else can use, that's all. OCM has a steady growth in users (http://openchargemap.org/site/stats), each one can edit locations and add photos and comments. Personally a check-in/comment from a user is worth more to me than any other type of 'verified' flag becuase it means someone (one of you guys) actually checked it and could have updated the listing if anything was wrong.

Regarding licensing, it's a necessary evil and we're really just 'copying' projects like OSM etc who have tread similar paths. I'd personally like a public domain license of some sort on the data and in the future it may be that we can re-license. We really need a copyright/licensing expert to work through that though. It's true that we mix licensing in order to be able to host data about a larger number of locations and to make our service appear more useful. I don't think anyone really 'likes' the 3rd party data, expect for the users. I myself do like the 3rd party data which has been provided under an 'Open' license. Include &opendata=true in your API call and you will be fine, if someone has an issue later on, just blame us.

Incidentally, if you edit an imported location which is not 'Open Data' please make whichever substantial changes you see fit, updating as much of it as is possible based on your own knowledge, it then becomes a new 'Open' listing based on your edits and we archive the old listing. The more edits that are made the more 'Open' the data becomes.

We need more people making edits and we need more people who are willing to become 'Editors' - those with time, competency and interest who can then have the power of immediate publishing and who can approve the edits of others (we do need to streamline our process).

Overall, It's not an easy thing we're doing, but we are trying as best we can with the resources we have.
 

·
Driving yet another EV!
Joined
·
2,578 Posts
I have just checked out the latest version of http://openchargemap.org/app/ and it has changed quite a bit! Unfortunately it is like trying to route-plan through a letterbox. On my PC the map is 1275px wide by only 420px high (it seems to limit to 420px high no matter what). PlugShare give me a map that is 1920px by 925px; it makes almost all of the screen available as map. I'm afraid I like a big map! Is there anything that can be done?
 

·
Registered
Joined
·
62 Posts
I have just checked out the latest version of http://openchargemap.org/app/ and it has changed quite a bit! Unfortunately it is like trying to route-plan through a letterbox. On my PC the map is 1275px wide by only 420px high (it seems to limit to 420px high no matter what). PlugShare give me a map that is 1920px by 925px; it makes almost all of the screen available as map. I'm afraid I like a big map! Is there anything that can be done?
Yes, that's no problem at all I'll get it changed soon - I consider the web version a 'beta' for the forthcoming mobile apps so it's certainly up for discussion as to how the UI works.

The main problem is choosing how to present things from all of the possible ways it could be done, at all screen resolutions, and in a way that broadly makes sense. If we do have any UI designers on hand I'm happy to have drawing board type discussions - nothing is set in stone, restrictions to do with dev time available applies as usual.
 

·
Premium Member
Joined
·
8,939 Posts
If we do have any UI designers on hand I'm happy to have drawing board type discussions - nothing is set in stone, restrictions to do with dev time available applies as usual.
That's me (mostly) by paid profession... but really don't have the time to give at the moment. So many things I'd like to spend more time on and help with!

Paul
 

·
Registered
Joined
·
62 Posts
That's me (mostly) by paid profession... but really don't have the time to give at the moment. So many things I'd like to spend more time on and help with!

Paul
Certainly understand that, I'm supposed to be developing a pest protection warranty database instead of replying to this thread. Oh the glamour of my profession.

Feedback is as good as contributing actual designs, as long as it's constructive :)
 

·
Registered
Joined
·
120 Posts
Following this with interest, and I use a variety of maps. Mostly plugshare as this works for me. But, I am also sending updates, photo's and new information to a variety of others to try to get them all as accurate as possible although this is time consuming.
I have the plugshare app and the CYC app on my tablet, and when I am charging and have nothing else to do I play around with them. I use the plugshare to check in, then go the CYC one and find directly that this takes 5-10 times as long. To load, to find places, to do anything. That is going to be the most important issue in my opinion, make it fast. Even if this means that you cache data that the user has as favourites, with a flag to indicate that the cache needs to be updated in stead of loading all data in each time. Most points will be steady information, if I can get some details and an indication that this station has updates I can then initiate the update if required. This will allow me to quickly scan my favourites and get the information I require.
 

·
Registered
Joined
·
582 Posts
@MetalHead For what it is worth, I would certainly support OCM if app-based data collection was introduced.

The fact the user can plot their exact spot (and or modify it using an aerial photograph) makes for excellent accuracy. There will always be some who either don't know or care about the accuracy (or, to be honest, spelling or useful info) but if someone can update it to correct means the data is invariably more accurate than (say) Zap Map or CYC who placed pins with cavalier abandon and requires emailing them - if anyone can be bothered.

Fingers crossed!
 

·
Registered
Joined
·
647 Posts
@MetalHead For what it is worth, I would certainly support OCM if app-based data collection was introduced.
We're working on this very thing through OVMS. One of the niceties about doing it through OVMS is that it can be done autonomously and will be quite accurate and reliable because the OVMS module, in communication with the EV, are reporting the information to OCM. Hence, you get accurate information about the available current, precise GPS location, and type of car used to do the charge.
 

·
http://ts.la/david3488
Joined
·
642 Posts
If Carwings mapping was pulling from ocm I would be happy and you could close them all down.
Im sure google/tesla/ocm will work well. test drive soon..
 

·
Registered
Joined
·
582 Posts
OVMS is fine for the hobbyists, but for the rest of us who are using EV's as an ICE replacement, this won't be any incentive. Also a good number of Chargepoints are in GPS-free areas (multi-stories, underground etc) where the circle of accuracy can be pretty loose.

Why restrict to the vehicle? A good number of sites I visit (but cannot charge from) are Rapids. I need T2. So even IF I had OVMS, you wouldn't get any data! I'm around 80% complete verifying charging places in Glasgow - I've visited every one and corrected and amended on PS. I plan to complete this by August before doing the ones in the East.

Don't restrict yourself to OVMS, as alone this may be part of the problem, not the solution!
 

·
Registered
Joined
·
7,249 Posts
I agree. OVMS is not the only solution but it could be part of the answer.

Not everyone will have OVMS and I think you are probably right to say that at the moment it is mainly enthusiasts that have it. However, for those that do have it it could be a viable way to verify/update/add charge location data where there is a signal. Also, OVMS is not standing still and it won't remain as it is today so there could well be a lot more functionality and options in the future.

So yes, it is and may remain a minority product but could still be a valuable aid.
 

·
Registered
Joined
·
528 Posts
LOL o_O Some of us would like to buy an OVMS, but they've been out of stock for weeks, with no further info etc. it looks like a dead project... Shame...
 

·
Registered
Joined
·
647 Posts
LOL o_O Some of us would like to buy an OVMS, but they've been out of stock for weeks, with no further info etc. it looks like a dead project... Shame...
There are posts on the OVMS mailing list almost every day. Some people are hacking on the code regularly, and git commits are happening often-enough that it's not dead. No doubt, however, that the project could use some more help.

If you want to build one from components, you can. The hardware is open-source and the components are readily available, I think. You could ask on the mailing list as to what is the hold-up on module availability, but I do know that there are plenty of messages on the mailing list about a new revision of the hardware... so maybe production is stalled waiting for a new build.

The current (or "old") hardware has an expiration date on it, since in the US 2G GSM service is likely to disappear as AT&T (the only US provider that I know of) will be killing it on 31 Dec 2016 (as announced). At that point the hardware will need to be modified to support 3G which will require a different processor, I think. It's possible that AT&T's announcement (which was recent) has prompted a halt to the "old" hardware production in lieu of a new hardware build that will be compatible with 3G GSM service.
 
41 - 60 of 70 Posts
Top