Being the co-founder and lead dev of Plurk I can tell you this sucks :-( But we are looking at our legal options, thought legal war with Microsoft and especially legal war at Microsoft in China would require a lot of resources.
I think this will become a major concern for all startups that are doing lots of client-side stuff. If people have determination they can copy your code (even if all your code is obfuscated, which Plurk's is). And I don't know what the solution is - other than stronger IP laws in emerging countries.
Was that really the prime motivator in joining MS? How do you mean "be a voice for startups"? Cool.
People usually cite this reasoning for joining the government -- an organization whose mandate is toward the greater good of its people, but I've never heard it applied toward a corporate entity, whose mandate is solely profit.
Would you mind posting a bit more about your motivations and thinking? I would really like to hear.
So Microsoft VPs want to find out why people do not accept their products. Could be a good thing, if they would then insist on improvement of the products guided by the data that the "dev evangelist" provides. You, though, make it sound like it's a bad thing. Care to explain?
I agree. Its highly unlikely that this was done on purpose with complete transparency to higher management. But I still wonder how the manager or the manager's manager of the project didn't see this happening. I am pretty sure the team and their 1 or 2 levels up researched plurk and its success in the region enough to recognize that it was copied. The product manager could not have missed this. I am sure a lot of people are under fire for this.
You are clearly in the right. But a bit of advice: skip the whole look and feel issue* and stick to copyright infringement. Their code is exactly the same or highly derivative of Plurk code. This is pretty well-worn copyright territory, I don't think it would take much for Microsoft to back down.
* Apple sued Microsoft back in the day for stealing the look and feel of Mac OS, and failed. So that argument can be very hard to make.
Under U.S. law, copyright protection automatically affixes to any original work of authorship that becomes fixed in any tangible medium of expression. It does not extend to ideas themselves.
Illegal copying can occur by exact copying of source or object code or by paraphrasing (misappropriating a program's structure or architecture, its inter-modular relationships, and/or its algorithms and data structures, at least insofar as these elements do not constitute unprotectable ideas - this can occur by copying the fundamental essence or structure of a copyrighted work even while rephrasing or restating it in immaterial respects).
Impermissible copying can also involve exact duplication of a user interface, or a "paraphrasing" of that interface (e.g., by creating screen displays and menu commands which are very similar but not identical to those of a copyrighted program).
Though interface elements can potentially be copyrighted, the advice you give is sound in most cases because of the courts' reluctance to give any party a monopoly on interface ideas in the guise of copyright (thus, Apple had sought unsuccessfully via copyright to gain exclusive use of a variety of interface elements, such as a visual depiction of a trash can). This is where "look and feel" cases become so difficult - who wants to give anybody exclusive rights via copyright, for example, to the basic ideas behind a spreadsheet? That is why many of those older cases failed.
In spite of this, I don't know that I would simply dismiss the look-and-feel aspect of this case based on the interface involved here. On its face, there appears to be more involved with this interface than mere ideas and, when there are creative forms of expression involved, and the copying is exact or so significantly "paraphrased" as to be almost indistinguishable, an interface infringement may have occurred.
All of the above, however, is based on a U.S. law analysis and may be irrelevant (at least legally) in China. If there is some way to bring U.S. legal principles into play, or anything analogous to them, then this would appear to be a fairly flagrant instance of illegal copying. While the costs of launching and pursuing a formal legal fight would be great, as long as you have a reasonable venue in which to launch it (i.e., one that respects IP rights) and the resources to carry it through to the preliminary injunction stage (usually 6 to 8 weeks into the case), winning a preliminary injunction would likely work its effect in causing Microsoft to fold, assuming it otherwise is willing before that to fight and is able to survive a horrible onslaught of bad publicity.
So, how does this work? Will everything be fine if they maintained the same look and feel and intentionally obfuscated/added more junk to their code to make the code look different?
Look and feel is really hard to pin down legally. It's similar to the fashion industry. You just can't copyright a design -- and that is why there are so many knockoffs.
But that's just how it is in the design industry -- new designs come out quickly, so it's usually not worth the effort to fight knockoffs.
You just can't copyright a design -- and that is why there are so many knockoffs.
That's because design is an area covered by trademark, not copyright. You can trademark a design if it's distinctive or innovative enough, at least in the U.S., and have protection from imitation. You don't even have to register a trademark to establish rights, you can declare your rights simply by using the (TM) symbol, although you do have to register in order to defend the rights in federal court or claim international protection.
From a whois check on their domain name, however, Plurk Inc. appears to be Canadian. Anyone have details on the what differences there are in the Canadian trademark system?
No. The Apple v. MS and Lotus v. Borland "look and feel" results were about the interface, not implementation. If you steal code (or other copyrighted content) from someone, no amount of obfuscation will "make it OK" legally (though I guess it might make it impossible to detect).
But simply cloning a user interface is legal, at least in US case law.
>But simply cloning a user interface is legal, at least in US case law.
Can you give us citation that it is legal, or are making your claim based on the above lawsuits?
Having a look and feel that is "similar" but "not the same" is open to debate, but having 90% of the interface ripped off, pixel by pixel can't be just "ok". We don't know that because something so dumb didn't happen before in the highest level.
The Lotus v. Borland decision was, in fact, about an interface that was keystroke-for-keystroke identical between the two applications. Again, there's a very clear case here for copyright infringement of the code. Mixing that with a much less clear case based on the status of the "interface" is just not a good idea.
Lotus v. Borland was not the same. As a matter of fact it was a very stupid lawsuit and rightfully dismissed.
>None of the source code or machine code that generated the menus was copied, but the names of the commands and the organization of those commands into a hierarchy were virtually identical.
In this case MS not only copied the interface of plurk, but from the look of it they also stole the code. Such a blatant copyright infringement has never happened before at this level to the best of my knowledge.
Your claim that it is "legal to copy UI in US law", based on the above example, is false.
Copying a menu, its command name and hierarchy is not the same as copying an interface. A menu is only small sub-set of user interface.
I feel for you, man. This is really awful to wake up and see.
China has totally different laws though -- do you know if this kind of copying is legal there? I have a number of friends who have told me that the whole web startup scene in Beijing is totally different than the US because the rules of the market are completely different over there (rampant copying, theft, DDOS attacks among competitors, etc). One example of this I've heard is that game developers have to give away their games for free and then rely on in-game purchases to make money, because there's no chance of actually selling software upfront.
What are your thoughts?
Best of luck fighting this -- hopefully this will just be a funny war story in a few months time.
I can't disclose much on the legal aspects, since I might say something stupid. A few months from now I hope the war story will be similar to David vs. Goliath, because that is our current situation - i.e. we are fighting vs. a billion dollar mega corp that has stolen most of our ideas and code and uses their power to come in a market where we are banned [China].
Indeed. See the related Steve Blank post and surrounding discussion here: http://news.ycombinator.com/item?id=981167. This is a little different since Microsoft is copying a well-known and already successful product, but the idea still applies.
Yes, but just because everyone at the company didn't decide to do this in unison, doesn't mean that the "company" isn't 100% completely responsible for it.
If the "VP level" at MS doesn't have enough control to prevent an entire department in China from ripping off a popular site and going live with it... then damn. I see no difference between that and Ballmer himself copying and pasting the code.
This sounds horrible. Microsoft shouldn't be blamed for this entirely though. Why wouldn't the team that created this think through before?
From what I know, people inside Microsoft make sure they don't get in to any legal trouble before they make use of any code or idea that's already available.
Anyways, this episode is a blatant disrespect for IP. Copying the code is something awful. I believe this case deserves the attention of authorities at Microsoft.
Yeah I know. And if they decide to make use of any 3rd party application or code internally, they make sure they don't unintentionally mess up with legal requirements of the software. This certainly is an isolated case which needs serious investigation.
And yet not uncommon. We (Django) had a rather frustrating mailing-list exchange with IBM folks who wanted to build a DB2 driver for Django; they claimed their corporate policies wouldn't allow them to look at the existing Django DB backends, or anything BSD-licensed, to gather information on the API.
Eventually they wrote an adapter; I don't have any idea how they got the API spec, since they swore up and down they couldn't look at existing code.
(and, to be honest, the initial released version of their DB2 adapter showed some serious issues which may be related to not understanding how the code was supposed to work)
I don't think it's an extreme approach. It would otherwise be nearly impossible to prove that Microsoft's code is not derived from GPL'd works. At least without revealing their code.
I disagree; "you shouldn't read Open Source code" doesn't fix anything much because it's impossible to find out if developers are sticking to it or not (so it doesn't decrease the burden of proof). It's just as impossible to prove that MS work doesn't include GPL code with or without the rule :)
A better (more effective) one is, surely, dont reuse restricted code.... :)
Microsoft actually tries to enforce the policy of "you shouldn't read Open Source code". For example, they block access to Sourceforge from the Microsoft campus.
That is totally false. There are no restrictions of any kind on the Microsoft network in general (apart from your standard virus/malware/firewall type stuff).
I've synced code for my projects from sourceforge/google code/github often.
I realise the policy is probably not as originally worded here: but I was taking it somewhat literally to point out the fallacy of it.
At some point we have to trust devs to do their own work. This should show that no amount of policy can stop people ripping off work! Whatever the intention.
Reading code is always good; there is usually a net gain as long as it is ethical (i.e. not copied)
By the way, what do you mean by obfuscation?
Because this is really weird,
they even copied the variables. So was the code obfuscated or just trimmed of white space. You should consider using the google closure compiler with advanced optimizations, although you will have το do some extra work το get your javascript compile in advanced mode.
Just guessing: he's got a layout where alt_gr+t produces a small tau and is typing quite quickly hitting space+alt_gr by accident. I know that's the reason some of my k-s at the beginning of a word are ĸ-s. (you can just miss it while reading later)
By any means: use the visibility you gain from this for your PR/advertising game! Now is the time for a smart play, get the public behind you (in terms of: target audience goes Plurk 'cause it's cool). Granted, this is not 'the solution but an opportunity you can manage to get a win out of this. (Remember, PG wrote the $$ he spent on PR was worth it.)
Microsoft should be held accountable for this, but assuming the blog is correct, I'm assuming that the deliberate ripoff (described on plurk's blog) wasn't a high-level Redmond decision.It was probably a decision taken at the junior management levelin Microsoft, China.
It will be interesting to see Microsoft's response, but suspending the MSN club service is probably the best immediate response, if plurk's blog is correct.
Precisely. This was a junior dev somewhere far away from Redmond making a terrible mistake. It's about to get thrown up to the highest level, and I can pretty much guarantee you that MSFT takes the site down.
It wasn't even that. It was a third party contractor hired by the Chinese company Microsoft partnered with to create "MSN China". So this was beyond just a hierarchical separation within the company, it was several degrees separated from the company itself. Only through contract is this even owned by Microsoft.
The JS code was obviously unpacked/prettified. All internal vars have shortnames like B, C, D, etc. Everything that interacts with the outside has long names, as in the minified plurk code.
I'm speculating, but I'd guess this was the fault of poor management of an external contractor or partner. We don't know for sure who actually developed the code, just who's hosting it. It's not like similar things haven't happened before:
I'd personally never heard of Plurk until now, and wouldn't have known myself that a contractor had ripped it off had I commissioned the work.
It's still an egregious violation, but I doubt this will even need to go to trial. Microsoft stands to gain little from alienating startups in this manner. If it comes to it, I'm sure aggressively shaming them publicly will be far cheaper than suing and eventually as effective.
No surprises here - knew as soon as I saw the story that it was going to be something created by a contracted third-party and that Microsoft/MSN would not have recognised it as a rip-off.
"According to a source familiar with the situation, the Juku application was created for MSN China, which is a joint venture between Microsoft and a Chinese company. However, the source says the application was not designed by Microsoft or the joint venture, but rather by a third-party Chinese vendor hired for the task."
Plurk should sue and win. If M$ could have had issues over phonetically rhyming domain name and take the kid to court, why not plurk taking microsoft into court for something serious. You just can't excuse yourself by saying someone in my company's xyz office is responsible.
Interesting how javascript is turning source code theft into copyright infringement (a civil matter), when traditionally it would be trade secret theft (a criminal matter).
In general, there are three factors that characterize a trade secret:
* is not generally known to the public;
* confers some sort of economic benefit on its holder
(where this benefit must derive specifically from its not
being generally known, not just from the value of the
information itself);
* is the subject of reasonable efforts to maintain its secrecy.
Incredible, I guess in their culture this is ok (or has become so commonplace as to be ok), but shouldn't Microsoft have some sort of measures against it in their own workplaces?
My guess is the Chinese team was facing strict deadlines and in the act of desperation they ripped off the code and layout from Plurk. It's outrageous and MS needs to straight up get sued for this.
Plurk has a very small team of less than 10 guys and in my eyes this is nothing less than MS bullying. This is not right.
I don't know if I can lay all the blame on Microsoft for this.
The dev's knew what they were doing but I doubt the rip-off percolated very high up. As a SW manager I'm not sure how I'd know that this was ripped off wholesale if I wasn't very familiar with the ripped off site.
Don't get me wrong, this is bad. But expose it to MS and let them clean house. I don't doubt that they will.
I'm fairly certain that the higher-ups probably weren't aware of this. However, the company is responsible for the actions of all of it's employees. I mean, sure, Balmer had no idea about any of this is, but that doesn't mean that Microsoft isn't responsible... even if it was done by subcontractors. They are responsible for what they publish on the web, just as anyone else is.
I know the web world is different, but a couple of years back nobody from my product group was allowed to install a competitor product and play with it as it would cause IP prior art issues. Most of the developers did it anyhow, they just would not admit to it. However, I did not know of a single manager who really did it. So I would not be surprised if the managers had an executive policy which did not allow them to see other products. This was at a three lettered big company and not MS though.
Plurk is blocked by the Great Firewall in mainland China, so the guys at MS China have a good excuse insisting they did not know it before, though I would guess they have VPN to get over with the firewall ...
Something like this did happen at Google China a couple of years ago: they released a Chinese IME that contained a dictionary taken from Sohu. Sohu demanded a public apology, which Google issued a couple of days later, along with a version that did not contain the stolen list.
Edit: for what it's worth, I personally agree with the theory that this was probably done by a local team that was under a lot of pressure but wasn't getting enough oversight.
As a Chinese guy, I have to say that the above comment is generally correct - there is fundamentally a different awareness of the concept of IP in Asia.
More frankly, this concept doesn't really exist.
It's not that the culture is more tolerant of ripping others off, but rather that when it comes to creative works, it's not seen as ripping off. Being able to take, adapt, and re-release the work of others, even for-profit, is a matter of course for the majority of the population. Many are not even aware of the moral issues surrounding it, and as far as they are concerned, none exists.
In all likelihood this wasn't so much a team under pressure and stealing out of desperation - in all likelihood someone just did it without very much thought... and it flew under the radar of those who know better.
That's generally true of the Internet tho'. So many photographers I know have had someone rip off their galleries, the usual excuse is just "I found it".
When you put it that way, it does make the arguments of software publishers sound much like those of other media companies struggling with IP management.
China has some odd laws in terms of how Chinese subsidiaries work from what I understand. It is isn't quite like opening a subsidiary in say, England. I think there is a rule about a local chinese partner company being involved where you get your employees through that chinese partner company. In the example I witnessed, it was difficult to dictate culture in the chinese office as they were working through an intermediary. Can anyone else with more experience on this comment?
There are two kinds of subsidiary in China for foreign companies:
- Representation Office
- WFOE: Wholly Owned Foreign Enterprise
In the case of a representation office, employees are hired through a Human Ressource company (eg: FESCO) but not in the case of WFOE (which requires quite a bit more capital to open though: around 45 000$ minimum invested over a period of 2 years in Shanghai)
I'm just starting here, so I'm not exactly sure yet of the limitation of FESCO contracts in term of how much of the culture you can dictate in the chinese office...
This reminds me of the Kayak/Bing Travel issue from July and similar incidents before that. It's easy enough to blame some contractors at a distant outpost but the frequency with which these things happen around Microsoft certainly leads to the conclusion that they are bad actors and poor citizens.
Farecast had a very different UI when MS bought them. The transition to the copy of the Kayak UI happened well after Kayak rolled theirs out and also well after MS bought Farecast.
The scenario/timeline you outline above is reasonable enough to assume - as is the idea that, as Farecast came out before Kayak, that Kayak was actually copying Farecast. What actually happened though is different. Farecast had a completely different model - they were banking on their ability to forecast whether or not you were likely to get a cheaper fare for the same route if you waited. Their original UI was based around that idea and it didn't go over incredibly well. After the Microsoft purchase and before the launch of Bing Travel they did a major redesign and got rid of most of the old Farecast. The old Farecast price predictor idea is reduced to one line at the top of the flights list on the page center.
I wrote about this and posted to HN shortly after Bing Travel went live and because of that I've been introduced to more information about the whole situation than I had when I originally brought it up. In a nutshell, the Microsoft team blatantly copied Kayak's UI and has pretty much gotten away with it.
Whilst this really does suck, I think it's valid to mention here that Microsoft have now said that the coding was 'done' by a 3rd party developer so all the comments that MS have done this are no longer valid.
Having said that though, someones head should roll at MS for not checking this out properly before go-live and I'm pretty certain MS themselves are ultra embarrassed and will try to solve this as soon as possible.
I hope that the Plurk guys get something worthwhile out of this ... maybe even more than the copious amounts of free advertising they're currently seeing.
Plurk's blog page doesn't like being viewed at 1024x768 in Chromium... the left sidebar cut-off by the left side of my screen and the top-right Plurk logo is cut-off by the blog post.
He does have a point (one which merits better than his -4 fate).
If your biggest asset is your UI, not only do you have to appreciate you have a tenuous competitive advantage but you should have a battle plan, both technical and legal, for when the theft invariably does happen.
They got lucky that it was a big operation like MS with a public reputation to maintain (laugh all you like) rather than some no-name. It's easy to put public pressure on the Apples, Google and MS' of the world.
I just pretty-printed the largest JS file loaded in their homepage. The result doesn't look obfuscated by any means.
Removing whitespace != Obfuscating
Don't take their claims at face-value, get your facts straight.
Disclaimer: I'm not claiming that obfuscating (or not) makes any difference with regards to MS infringing on their codebase, just pointing out that it's not actually very obfuscated as claimed by their developer.
I just re-read the article. If you really squint you might be able to assume that their code was scrambled in some way (if that's what "not compiled" means). Otherwise if you're trying to gain (well deserved) sympathy for something like this then the deobfuscation process is probably the most important part of the story. Did MS then rescramble the code? I'd want to know that as well.
Of course it has context if this is being presented as a legal issue. The original article doesn't really mention it (although it's easier to assume now with the hindsight of a complete comment thread). I believe you are "even more guilty" if you de-obfuscate someone's code, then copy and republish/reuse.
OK - I can see what you mean. However, I still disagree. De-obfuscation does not provide more evidence that infringement occurred, it simply implies that whoever did it likely knew they should not be doing it, which has (I believe) absolutely no legal implication.
Not condoning the crime, but does anyone else find it ironic that a black guy is accusing someone of robbery?
Yes, because ignorant stereotyping is fun! Remember kids, it's not racism if it's not targeted at a member of the List Of Races We Should Be Sensitive About!
uhh.. its not stereotyping - comment was directed at a government policy that has fostered piracy at all levels in China. The comment is a characterization of policy, not a characterization of people.
I can't say I'm surprised that Microsoft has stolen yet another idea, but I am astonished that they didn't go to more effort to obfuscate that fact. I'm publicizing this through my networks. More people need to know that the 's' in 'Microsoft' is really a '$'.
I've always found the use of $ when referring to Microsoft to be really really childish, and weird when I see people here, who are presumably mature adults, doing the same.
I think this will become a major concern for all startups that are doing lots of client-side stuff. If people have determination they can copy your code (even if all your code is obfuscated, which Plurk's is). And I don't know what the solution is - other than stronger IP laws in emerging countries.