In Israel, Class-Action Plaintiff Requests Waze Source Code Under GPL

jonklinger (1166633) writes "A class action lawsuit was brought against Waze (a community-based traffic and navigation app), claiming that their source code and map data were licensed to Waze by the community under the GPL. The plaintiff, Roey Gorodish, requests a copy of the recent source code and map data. This is (as far as I know) the first ever GPL class action suit, too bad it will be quashed by bad facts later as I see it." Google seems to do a credible translation of this source article.

Comments

Re:The Googles Translation

[interkin3tic]

Google owns waze.

What basis for this case?

[gnasher719]

So is there any code in Waze that isn't now owned by Google and that was given to Waze under the GPL license? Can we see the exact terms and conditions under which users give data to Waze? You can make all kinds of claims, but eventually you have to come up with the goods (unless you are SCO obviously).

Under US law you have no right to the source code. A copyright owner may have the right to sue for copyright infringement. We'll see how this is different in Israel.

Re:What basis for this case?

[Jane Q. Public]

"Under US law you have no right to the source code. "

Ahem... if Mr. Gorodish is correct, and Waze was licensed under GPLv2, then we do in fact have a right to the source code, and Google would be breaking the law by not providing it.

GPL v2:

3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following:

a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

b) Accompany it with a written offer, valid for at least three years, to give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or,

c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above.)

. . .

6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License."

In this particular instance, "you" would mean Google. If, that is, Waze was GPLed.

Re:What basis for this case?

[Richard_at_work]

Actually gnasher is correct - you have no *right* to the source code, you only have a *right* to either insist they abide by the licence terms or be in violation of copyright, and if its the latter then you need someone whose copyright is being violated to bring suit against them as no one else can do it in their stead.

Re:What basis for this case?

[gnasher719]

Ahem... if Mr. Gorodish is correct, and Waze was licensed under GPLv2, then we do in fact have a right to the source code, and Google would be breaking the law by not providing it.

No, you don't have the right to the source code. _If_ there is GPL licensed code and the code is licensed under the condition that Google should give you source code, then Google has the free choice of complying with the license (giving you the source code) or committing an act of copyright infringement (not giving you the source code and having no valid license).

You can't force Google which of these choices they take. And if you are not the copyright holder, then you have no standing to sue. Even if you _are_ the copyright holder, you can't force Google to give you any source code. You could say "give me the source code or I'll sue you for beeeelions" and they can say "Ok, sue us", lose the case, pay billions and keep their source code.

Re:What basis for this case?

[jbolden]

Jane --

Waze owned copyright on most (if not all of the code) they weren't a licensee. Which means they can relicense at will. Now of course any code released under the GPLv2 by them is still under GPLv2 but that's it.

Re:What basis for this case?

[gnasher719]

Actually gnasher is correct - you have no *right* to the source code, you only have a *right* to either insist they abide by the licence terms or be in violation of copyright, and if its the latter then you need someone whose copyright is being violated to bring suit against them as no one else can do it in their stead.

Thanks :-) Importantly, I read that an early Waze version was GPL v2.2 licensed, while a new one is proprietary. The question is not just how was the first version licensed, but who owns the copyright. If Waze wrote all that code themselves, it doesn't matter what license, because only Waze could then sue themselves for copyright infringement.

Re:What basis for this case?

[Kjella]

Ahem... if Mr. Gorodish is correct, and Waze was licensed under GPLv2, then we do in fact have a right to the source code, and Google would be breaking the law by not providing it.

Yes, but you generally can't ask for the contract to be fulfilled in court. If you agree to build a house for me and fail to fulfill the contract they'll award damages but they won't drag you there in chains to build the house. There are a few narrow exceptions to this called "specific performance" involving unique real estate, antiques, works of art, collector's items and such where the court may insist the transaction be completed, but it's very much the exception and only when money can't fully compensate your loss. They'll never force a company to open up their own source code, even if they somehow managed to mix it with GPL code. Both sides might of course offer it as a settlement or parts of one, but in an actual court verdict it will be a dollar value while the proprietary source stays closed source.

Re:What basis for this case?

[msauve]

"if Mr. Gorodish is correct, and Waze was licensed under GPLv2, then we do in fact have a right to the source code, and Google would be breaking the law by not providing it. "

Only if it contained code not written by Waze (or Google, who is now the copyright owner). If you write a program, and release source under any version of the GPL, you're perfectly free to not release source for any future versions - you're the copyright owner, you're not relying on the GPL for your rights. The GPL terms are effective on _others_, who may modify the source and are then required to release the modified source when the distribute their version.

The attempt to tie map data into the situation is like trying to claim that a word processor must be open sourced if it is used to edit open source code.

Re:What basis for this case?

[HiThere]

But you could sue them again for every time they "distributed" it after being notified.

I suspect that the code is clean in one of two ways:
1) Portions were never under the GPL, and
2) Portions were written by the entity that sold them (was sold) to Google.
There is also likely a "de minimis" amount of code that is unclean, and is also not "functionally required", and thus coverable by copyright law. Being "de minimis" (i.e. minimal in quantity) it will probably be officially forgiven by the court system.

N.B.: This is suspcion, not knowledge. But Google has a lot of experience with GPL code, and usually plays it straight. Being under the GPL doesn't automatically mean it must be public. That's true only if the author insists AND it is publicly distributed.

P.S.: I suspect that much of this code is never distributed by Google, and the GPL is not the AGPL, so server side code doesn't need to be distributed, even if it is covered.

Re:What basis for this case?

[Kalriath]

Actually they couldn't, as due to the fact that you do not need a license to use your own code, it is impossible for Waze to infringe their own copyright. Which is to say, if they wrote all that code themselves then noone has any grounds to sue.

Re:What basis for this case?

[Kalriath]

You can't. The copyright holder can. Unless it's themselves, in which case no infringement occurred.

Re:What basis for this case?

[Jane Q. Public]

"Actually gnasher is correct - you have no *right* to the source code, you only have a *right* to either insist they abide by the licence terms or be in violation of copyright, and if its the latter then you need someone whose copyright is being violated to bring suit against them as no one else can do it in their stead."

If you want to get into hair-splitting mode, then yes. But I think that's going a bit far.

Copyright and patents are the only "rights" mentioned in the Constitution that are bestowed by government... and those only temporarily. All other "rights" are presumed to have already existed; the Constitution merely acknowledges them, it doesn't "give" them.

So if you want to compare copyright to what most people consider to be Constitution rights, even if the document uses that word, it still isn't the same.

So again: if you really want to get into extremely fine hair-splitting, then copyright isn't a "right". But the Constitution does at least call it one.

Re:What basis for this case?

[sabt-pestnu]

> You could say "give me the source code or I'll sue you for beeeelions" and they can say "Ok, sue us", lose the case, pay billions and keep their source code.

... and be charged with contempt if they continue to distribute without a license.

And aside, it seems likely that part of the judgement would include "cough up the source code, mac."

problems

[jbolden]

The article linked is really quite good. According to the article are 2 prongs to the suit the code and the data.

Waze version 2 was GPLv2. Wave version 3 was proprietary. Waze claimed it rewrote the code. Roey Gorodish wants to examine the code to prove there was a violation but Israeli law doesn't allow discovery without some evidence. As an aside I also have question in that as as I know if Waze owns the code they can multilicense it. Roey Gorodish IMHO would have to find his code not just any code.

There is also the question of map data. Freemap’s data was released both proprietary and GPL and then only proprietary. I'm not sure what the basis of the claim is.

Re:problems

[Anubis IV]

Great summary.

As you, I have doubts that they have a leg to stand on (disclaimer: IANAL). Assuming Waze is the owner of all the source code, then no one else would have a claim to it and they would be free to license or re-license it as they chose, regardless of whether or not they rewrote it when going from v2 to v3. There's no obligation for them to keep it GPL'd forever, so if they wanted to take it proprietary, there's nothing stopping them from doing so, at least inasmuch as the code is concerned.

If they accepted code contributions from others that were licensed under GPL2 (which I don't think they did), then the question becomes one of whether or not they actually rewrote that code when going from Waze v2 to Waze v3, since leaving it in there would mean that they could potentially be in violation of the GPL, but if they never accepted any outside contributions and all work was done for-hire, then it would seem that they'd have every right to close it up.

The data, as you said, is a separate matter, but I'm a bit confused about the issue as well. The basis for the claim seems to be that Freemap promised to make all user contributions freely available, so because Waze used Freemap's data, it has a responsibility to uphold that promise as well. I can't find the terms of the agreement in English, but from what it sounds like, Freemap's data, which Waze used, was dual-licensed under the GPL and a proprietary license. The Google translation of Freemap's ToU seems to suggest that commercial entities were expected to procure the rights via the proprietary license, so I would assume that Waze used the proprietary license. As such, any of Freemap's code would not stand in the way of Waze going closed-source, and since user contributions to Waze would be governed by Waze's ToU, not Freemap's ToU, and would not necessarily be contributed back to Freemap in the first place, I can't see how the plaintiff has a leg to stand on. Waze would have every right to use Freemap's data as a starting point under a proprietary license and then accept data contributions on top of that without passing those contributions back to Freemap.

TL;DR: It sounds like regardless of whether a code rewrite was done or not, there isn't any GPL code in Waze anyway, meaning they're under no obligation to release it, nor is Waze under any obligation to pass user-submitted map contributions back to the source (Freemap) from which they got a subset of their data.

Re:problems

[jbolden]

Exactly. And just to add. Even if Waze did accept submission for version 2 that are still in version 3 . Roey Gorodish would only have standing if it were his code. This would be a copyright violation not a license violation and since he doesn't hold copyright....

I'm sure Waze had a proprietary license for the data they clearly said over and over they did.

This seems like a poorly thought out case. . Roey Gorodish loses quick IMHO.

Re:problems

[Andy Dodd]

One question: Is Waze v2 a derivative work under the GPL, or an original work that they happened to release under the GPL?

If the latter, then as the original copyright holder they are allowed to also release the code under an alternate license (assuming that all external contributions were either removed, had copyright assigned to Waze via a CLA, or had additional rights granted to the contribution via a CLA similar to Canonical's Harmony CLA...)

Also, as you've indicated, the only person with an actual legitimate claim in such a lawsuit is a copyright holder. For example, if you or I don't have any code in the Linux kernel that we retain copyright to, we can't sue someone for GPL infringement of the Linux kernel. (This is why kernel GPL violations rarely reach a lawsuit, most companies solve the issue well before a lawsuit or are located in China outside of the reach of most jurisdictions that someone could sue in, plus most kernel copyright holders would rather keep coding than spend time on a suit. The busybox team, on the other hand, frequently goes after busybox GPL violators.)

Re:problems

[jbolden]

An original work that happened to be released under GPLv2. There is some question whether any code was ever accepted into Waze outside of that owned by the group of people that signed all rights over to Google (i.e. if Google owned every line of Wave version 2). According to Google they unquestionably own ever line of Wave v3.

So yes you are understanding this properly and IMHO Roey Gorodish is going to lose fast.

Can start with symbol hunt

[SuperKendall]

Even without the source you could do a symbol scan of the app executable looking for unique method names in some target code.

I believe you have to jailbreak the device to get to the unencrypted application executable though.

Re:Can start with symbol hunt

[jbolden]

If the Waze founders own it then Waze may own copyright and not be a license holder which means they can freely relicense.

WTF if Waze?

[taiwanjohn]

Would it be TOO much trouble to mention in TFS that Waze.com is a "community based traffic and navigation app" and save me the trouble of searching it on Google?

Re:WTF if Waze?

[Soulskill]

Not at all -- I've updated the summary to include that.

Re:WTF if Waze?

[interkin3tic]

It's basically google maps with a different skin, and you can report things like speedtraps and accidents. And you get points, so there's a game aspect I suppose.

That's it. It's one of those "tech" companies that seem absurdly overvalued based on how little they actually do. In no sane world would it be worth the billion google paid for it. And on top of that, although the interface for reporting stuff is designed to be as minimal as possible and they prevent you from typing and driving, there's no way it's safe to use. I've used it, so I'm a hypocrite there, but it is a driving hazard.

Re:WTF if Waze?

[rsborg]

It's basically google maps with a different skin, and you can report things like speedtraps and accidents. And you get points, so there's a game aspect I suppose.

That's it. It's one of those "tech" companies that seem absurdly overvalued based on how little they actually do. In no sane world would it be worth the billion google paid for it. And on top of that, although the interface for reporting stuff is designed to be as minimal as possible and they prevent you from typing and driving, there's no way it's safe to use. I've used it, so I'm a hypocrite there, but it is a driving hazard.

Riiiight - just a map overlay and gaming aspect. No mention of incident reporting, showing aggregated speeds of waze users, the ability to share your location/trip,nope. None of that takes coding or infrastructure I guess.

Waze is more useful than google maps, apple maps, or navigon for me in estimating my commute times - it simply works. It even noticed a traffic disruption that had only really lasted for about 1 hr, showing me an alternate route through residential streets such that I got to my freeway onramp with only a 5m loss rather than what I guess would have been much longer.

I don't like that Google bought it - I don't trust Google very much, but I can't stop using Waze (I kill the app when I get home/work).

Re:WTF if Waze?

[foxharp]

it's not really like google maps -- at least, not the data in question. in israel, at least, the maps were 100% generated by Freemap, based on data contributed by the users as they drove the roads. (israel is a small enough country that you can pull that off.) the location and path data was uploaded from the client phone apps (which ran GPL'ed code, based on RoadMap) to a central server (which was always private code). so what's interesting is what the initial terms of service for the user-contributed map data was. i'd bet that this is like the CDDB case, where they were free to take the user-contributed CD database private at some point. as for RoadMap -- i'm a principal author of RoadMap, and am still the underworked maintainer of the mostly inactive sourceforge project. i looked into the GPL issue when the waze complaints started, and convinced myself at the time that certainly none of my code was in the current client. and frankly, i'd be shocked if any of the original code was left at all.

Re:WTF if Waze?

[interkin3tic]

What? I did mention incident reporting. In that very sentence you were responding to. A comma doesn't mean "stop reading here."

Re:WTF if Waze?

[msauve]

"It's basically google maps with a different skin"

It was developed by a different company, has a completely different code base, uses different map data, was created to monitor traffic flow and reroute users based on that, shows other users, allows users to map new areas by driving them, allows users to edit maps, and more.

But sure, other than that it's just a re-skin, in exactly the same way that Android is a re-skin of Windows.

A far more interesting story

[phozz bare]

I believe a far more interesting story about Waze has eluded Slashdot:

Two Technion students reverse-engineered Waze's method for detecting a traffic jam, then created a network of fake clients that reported traffic patterns that caused Waze to mark as jammed what was in reality a perfectly empty road.

Sources: Jerusalem Post, Wired.

Re:A far more interesting story

[Overzeetop]

Kind of a non-story, really, don't you think? How hard is it to believe that a crowdsourced data application can be poisoned by an intentional attack using spoofed data?

cf "man strip"

[rewindustry]

i doubt your theory would have much practical application here..

Re:The Googles Translation

[DrXym]

Waze has one good feature - it's far easier to cache maps for offline use. Happens automatically just by browsing around while online. In Google Maps you have to find the option and do it manually. Some of the "social" aspects of Waze seem pretty retarded though. I'm supposed to be driving a car, not "liking" other drivers or their traffic tips.

Re:Oblig:

[Minwee]

Are you saying that 10,000 people just heard about Waze today, or that 10,000 people just discovered the power of lazy copy writing?

They want the DATA?

[bobbied]

I get the GPL issue. If they took GPL code and modified it for their own use, they may be required to provide source, but what I don't get is the data. How are they justifying asking for that? Most "data" of that type that I'm aware of is licensed under the Creative Commons license (or similar).

For mapping applications, the *real* value is in the data. Where much of the data is freely available in raw form, there is significant effort required to process the data available and put it into a format to be doled out to handheld devices in small chunks. I don't see how they can claim that GPL entitles them to the raw data from Waze.

Re:They want the DATA?

[jbolden]

They are claiming the particular data was licensed under GPLv2. However the copyright entity Freemaps have never claimed that Waze's proprietary version was a violation. Waze claimed to have a separate license and they've never disputed this claim.

You don't know Objective C, do you?

[SuperKendall]

Objective-C is a message passing language. As such, it needs the message names you are sending to exist. They cannot be simply stripped out, they would have to be replaced which is a process that no-one I know of goes through.

Even if you strip debugging symbols, the names of the methods are all still there.

You can dump the class info for all of the Apple frameworks, do you really think they send those out un-stripped?

Yes, but who does that?

[SuperKendall]

Wouldn't simply obfuscation techniques defeat this?

First of all, they are not simple - you can in theory replace all of the existing message names in a binary with random text, but then you'd ALSO have to deal with code that manually calls messages expecting the name to still be there.

As a practical result, pretty much no-one does this. I'm pretty confident Waze would not have bothered - what would be the point?

Israeli Here...

[Anonymous Coward]

A few articles on this. First of all, why not link to the actual article rather than this guy's blog?

http://www.themarker.com/technation/1.2281212#

Google translate works decently well, but I am happy to translate if anyone cares.

Secondly, as someone who knows some of the parties on both sides (directly and indirectly - Israel is a "small" place), I can tell you this claim has a lot of credit from what I have seen and heard so far. The founders of Waze clearly were after a money grab for a long time and would do anything to get it.

Regarding Waze itself, the app became popular in Israel mostly because everything else including a lot of GPS devices had the worst maps of Israel you can imagine. Waze built its entire business here in Israel (where it took off initially and spread to the rest of the world) on the basis of having better maps than everyone else. That was always the #1 thing they said and selling point, so to use other people's code and data is very serious here. As for the app itself, it had huge problems with zooming, spamming, and all kinds of things. I think everyone was a sort of reluctant user if not for the maps. For example, people love spamming police still, putting stupid things like "snow ahead" in the middle of the summer which is obviously moronic in most of Israel, and so on. Not to mention the map itself becomes a litter of icons and stupid things if you don't turn tons of things off, and it eats your battery.

The idea of "social" mapping and a map out without any real possibility of a revenue stream that won't piss off users is ridiculous. I am shocked Google paid anything for it other than to crush it, and yet they are hiring heavily for Waze in Israel right now. They have always been a pretty cocky company here and if the allegations are true which it seems, they deserve everything they get and more.

Re:total contradiction in terms

[bws111]

So GPL (which relies 100% on copyright law for its very existence) is anti-copyright? Explain.

Wrong headed and wrong legally

[msobkow]

The fact that you once released a product under any particular license does not require you to continue to release the product under that same license. As the copyright holder, you're free to change the license at any time.

Have they pulled the source code for the GPL version? Is it no longer available?

If it's still available, they haven't got a case to stand on.

Re:total contradiction in terms

[TheCarp]

Well he is correct on that, but draws the wrong conclusion.

The GPL is anti-copyright. The entire purpose of the GPL is a hack on copyright to use copyright law to enforce a set of terms that are a direct assault on copyrights.

However, that is why its right that this attack be in court and makes perfect sense. The GPL is a hack that uses copyright law against people who try to jealously control works they derived from GPL works. How does it use that law without using it? This is the entire point of the GPL right here.... to give community contributers a way to put pressure on proprietary software makers who want to benefit from the community and not give back.

GPL is anti-copyright because it creates a community within which copyright law doesn't matter. As long as you have no intention to release binaries without source.... you are entirely unrestricted. You play fair with the community, you have blanket license to do as you will.

You cross that line however, and you are not part of that community. You are a parasite, and will be made to play by the rules outside that community. The rules like copyrights.

do pay attention...

[rewindustry]

what value would the method instances be?

surely any sensible rewrite would export the identical methods?

how is this any clue as to the code behind the methods exported?

i call hot air, your posts are misleading.

Re:do pay attention...

[SuperKendall]

surely any sensible rewrite would export the identical methods?

Note quite sure what you are saying here.

Code that is using performSelector (not uncommon) would break if you re-write the method names in the executable.

If you are talking about new methods created as part of a method re-write, it's unlikely they would be exactly the same

how is this any clue as to the code behind the methods exported?

If you have unique method/class names (likely) and you see them in the executable, you have reasonable grounds for claiming your source has been used.

i call hot air

Keep digging.

i stand by my conclusion

[rewindustry]

the plaintiff in this case appears to be going after "damages" - as if his right to make similar "jealous" profit from his work had been contravened.

however, under GPL - by my reading - the plaintiff has no right to claim "damages" in the "shekels" sense at all.

therefore i claim he makes a mockery of the whole point of the GPL.

he should be suing to have the V3 source released to the public, not to be released to him in order that that he can pursue his claim for a share in the "profits".

Re:i stand by my conclusion

[david_thornley]

I don't know Israeli law, but in the US there is no legal way to force somebody to hand over the source code.

If I have written something and released it under a GPL, and a company distributes it (likely changed) without my explicit permission, there are two possibilities. One is that the company is abiding by the GPL and releasing its source. The other is that the company does not have a valid license and is in violation of copyright. Normally, the company has not made a specific commitment to abide by the GPL. (I've never made a specific commitment in the case of F/OSS; I just do what the license says.)

Therefore, the only way I have to legally enforce this is to enforce my copyright. This involves asking the court for an injunction on further distribution and/or asking for damages for the copyright violations already done. The company has not agreed to release source code, and that's not a legitimate remedy for copyright infringement.

What I would most likely do is offer to drop my suit if the company starts complying with the GPL, but I can't legally force the company to comply. Getting the injunction and asking for damages are the only ways I can make it potentially worthwhile for the company to comply.

Re:total contradiction in terms

[bws111]

OK. Nirvana has been reached and there is no copyright. How, exactly, are you going to compel anyone to release the source of anything? GPL, and its goals, are entirely dependant on copyright. Claiming otherwise is disingenuous.

Re:total contradiction in terms

[TheCarp]

The GPL is entirely dependent on copyright yes. Without copyright, it has no point. In the absolute strictest of navel gazing senses, this is true.

However the GPL was written as a hack on copyright. The goal of the GPL is to foster a community of free software develpers who share code openly. The GPLs goals do not require the GPL, it is simply an attempt at forging such a community within the current environment without giving quarter to the enemy...that is...without helping the proprietary developers, who control so much of the software in use.

Yes without copyright, there would be little recourse...however, it would be worst for them, who would have no recourse against people who share their binaries. It would put all the onus on them of jealously protecting their own code, with the full risk of not doing so on them.

completely disagree

[rewindustry]

is the algorithm that counts, not the name you stick on it - you may as well try patent the word "glass" - and enjoy the karmic consequence.

i see your point

[rewindustry]

thank you - i may not like what you are saying, but it makes sense.

"you'll like this.. not a lot, but you'll like it" -- Paul Daniels