So, both sides have pretty much spoken, and what irks me the most is that the LayerVault guys haven't provided any concrete evidence to support their claims; they've posted vague responses and joked about DMCA'ing their friends on Twitter. There's also the question of why they would even pursue something like this. A simple Risks/Benefits analysis would've surely led most people not to issue a DMCA over a few icons they felt were stolen.
It's hard not to lose a little respect fo the LayerVault guys, even if the icons in question were direct copies. I get you're protective of your art, but this doesn't really make sense from a business standpoint, and their response has been less than convincing. It's really disappointing.
I could understand a DMCA if the icons were direct copies and if FlatUI had not responded to any email requests.
Even then it's a good idea to put up a quick and easy comparison sheet.
But for things like "3 cogs" (cogs are different sizes, different colours, different numbers of teeth, different orientation) it's ridiculous to DMCA it. 3 Cogs is so generic that it's an annoyance to engineers. (Because the cogs are usually drawn in such a way that they cannot possibly turn).
HN is generally pretty good when people rip off a design. Even people like me (who have very little idea about design) respect the amount of work and expertise that goes on. But it's because I have so little clue about design that people need to explain the similarities. If you're the first people to curl a newspaper under then tell me that. Show me what people used to do, and how you innovated that.
Is this the best example of the "infringement"? If that's the best they got than Layer Vault are looking pretty foolish (insecure, full of themselves, etc). What a joke.
Yes. Many people supported dcurtis and were angry at the blatant rip off. Also, many people were annoyed at dcurtis and felt he brought it on himself by acting like an idiot.
An HN member created a new blog thing; invite only; very designery. They were trying to build a brand for great writing. The way it was presented to the HN audience was sub-optimal - "Here's the great new shiny thing. You can't use it, you're all stupid."
Someone very quickly ripped it off.
There was intense discussion. Some people felt it's fair game to rip off the design. Other people felt it was a poor copy. Others were angry that it was such a blatant rip off (especially because it was so bad).
I've tried to describe this as neutrally as possible, but I understand it will sound pretty biased.
I believe obtvse at the time of the dispute was more directly ripping off svbtle than it is now, but I could be misremembering.
The unintentional pretentiousness of svbtle was amusing at the time and I can remember the urge to create a clone and replace the dammned kudos hover trap with a fontBomb[1].
Hello.
The way that works, you have to infer the existence of A for the existence of B and the existence of B for the existence of A.
Direct copies only work as an argument if the item is an Original, as in the idea of a thing wholly onto itself without an external influence of any sort.
you might argue that it's derived as in Originated from, borrowed from, a sourced to, an antecedent of, an allusion towards. I could continue.
sort of like how Star Wars Episode II and Star Wars Episode I are derived from Star Wars Episode IV. :)
It would actually be a derivative of ep VI, since the characters and creative designs differ and precede it.
A direct copy, would be the _exact_same_ characters as depicted in ep VI, but alas they are not, they are however similar to, which makes it a derivative of.
Judges, the people who arbitrate copyright disputes in the end, are humans, not computer programs. I don't think that if you tried to rip off Yoda, but made slight alterations to his face wrinkles, that you're going to convince a judge that it's something other than Yoda. A judge is not going to see Yoda as anything other than the same character through all six movies, no matter how many differences between the 3D model in the Phantom Menace and the 3D model in Attack of the Clones a "diff" command might find.
You can lose your rights if you don't protect them. Given this, businesses pretty much have to protect their work, or they'll run into bigger problems later.
If you'll indulge a little contextual Show HN, I made a single-page site that lays out the differences between the three major IP law domains: http://iplaws.info
LayerVault, the same people who brought you DesignerNews.
If that's indeed them who sent the notice, they are just shitting all over their reputation and alienating a lot of people. Talk about "starting off on the wrong foot"... and for what? A design that is border-line trivial. It would seem that their best option for getting the situation under control is to release their own version of the toolkit. A version that is sufficiently different from their original background-color, border-radius and icon @font-face. If not, they are bound to repeat the clone wars of Svbtle.
Edit | LayerVault actually appears to be a pretty useful product, clearly with a lot of thought sunk into it. Which makes it twice as unfortunate that they decided to pursue something like this. Google "LayerVault videos" if interested.
I lost respect for LayerVault when their software that I paid what I thought was good money for totally corrupted the PSD of an entire project that had 2 months worth of work put into it.
I thought I had a backup on my laptop, desktop, remote backup from my desktop and all the revisions on their server. But all revisions were corrupted and it synced those bad copies to all my machines, including the machine that performed it's own remote backup (so all those backups were broken).
Their support team acknowledged that it was a problem and that they were 'hoping' to fix in an upcoming version, but offered absolutely no solution to fix my data except their apologies.
Never going to trust that product ever again if they don't have data integrity as their number 1 priority.
This is bad but it also looks like you didnt have a proper backup strategy. Their service looks like a versioning/syncing service but not a backup service.
Well that's the thing, I actually thought my backup strategy was decent enough. Multiple versioned copies in multiple places.
My desktop was being backed up nightly to a remote source. Though it was the laptop (which wasnt being backed up) which I was doing the work on / changes to. But I saw that layer vault was syncing the changes to the desktop so figured it was fine.
So it went Laptop -> Layer Vault -> Desktop -> Remote Backup. But of course everything that passed through Layer Vault got corrupted. And eventually it 'resynced' all that corruption back to the laptop and the file went completely dead. So there ended up being no good copies of it.
So yeah in hindsight can say it was my fault for not doing nightlys of the laptop. But at the time I thought my strategy was fine, trusting that while at worst I'd lose a few versions or something if something terrible happened, not for it to actively destroy the file.
One kind of anything is a single point of failure.
Well hindsight is always 20/20, but your backup strategy still relied on a single service (that wasn't your own). Backup is one thing no one should completely rely on a third-party. Whether it's a USB drive, good ol' DVDs or what have you, anything else "of your own" is crucial at least weekly if not end-of-day.
The effort going into your backups must match the value you place on your data.
no the problem here was not checking the integrity of your backups.
not a "one kind of anything" single pt of failure problem. he expected LayerVault to not corrupt the PSDs sent through them (and the syncing actions made it worse), which is not unreasonable. it's the same as trusting Photoshop to write PSD files to disk that are identical when opened later.
the difference is that Photoshop is more time-tested and well-known, which changes the odds, but in your argument still would make Photoshop's saving mechanism a single point of failure as well.
and then what, use more different graphic design tools? :)
in your example, if you burn your backups to a good ol' DVD, don't you check that the files are actually on there? and have the burning tool check the integrity? and finally, if you don't check a few of those PSDs to actually load in Photoshop, who knows that the data your DVD burning tool received was correct?
You're right, backups are meaningless if they're unchecked. No different than dumping to a tape drive that's never verified and you get weeks of... nothing.
yeah could have easily happened to myself, things like that happen. What i dont understand is why your Remote Backup Source has only one (the latest) version of the file ?
Ah because the Remote Backups were being done of my Desktop (the changes were being done on my Laptop).
As the Desktop was getting the changes via the LayerVault syncs (simliar to say a Dropbox folder) each sync from LayerVault was sending a corrupted file.
So there were plenty of 'versions' of the file in remote backup, except these were all totally corrupted!
It's understandable to think that, but look at it this way: If Git did this sort of thing Hg would have won that battle hands down. I realize designers don't have any alternatives and this is one of the first solutions out there, but it is really important it be well built or at a minimum that they, LayerVault that is, provide a solid backup strategy to ensure they protect their users.
Designmodo announced Flat UI Free v.1.1 is public Now! Download updated version - http://designmodo.com/flat-free/ PSD & HTML (Twitter Bootstrap based)!
You really prepared to piss off hundreds of people by sending a DMCA takedown to each and every one? Really? Do you know when to stop digging? Is the hole deep enough yet? Found some suspicious odorous dark substance? I can tell you, it's not oil.
I agree on that. I mentioned the colors, becaue the parent comment said that it was only about 3 similar looking icons. The DMCA takedown request senders thought there's more to it.
It's true that all these flat UI themes are really similar, so It's really interesting how this will turn out and will the Flat UI repo will be brought back.
However in order to win an infringement claim you'd have to show that they constituted a substantial part of a work and that the colours were copied and not derived from a third party source or coincidentally the same. The usual civil law test is the balance of probabilities.
I think a more likely claim is that the matching colour scheme were a trademark infringement, an unlicensed use of trade dress (jurisdiction dependent of course). AFAIK DMCA takedown notices don't work for trademark infringement though?
If we look at Webdings (http://en.wikipedia.org/wiki/Webdings - TrueType dingbat font developed in 1997), there's actually a newspaper icon there that looks a lot like all of these:
Thanks for posting this--I hadn't seen the side by side comparison. If this is what's being claimed as 'copies' of their artwork, then it's clearly an empty claim. The Flat UI icons look better, to my taste.
So it seems that you have to provide exactly zero proof and your word (in a good faith) is enough to take a repository down. There is no way to verify anything and in order to stay safe they (github in this case) have to take the content down.
Now imagine I file a takedown notice for every aspiring project posted here on HN that is hosted on github. They take a pretty good hit from it I'd assume, to be silenced on the launch. Of course I open myself to litigation with false takedown request... but what do I care? I'm not a citizen of United States and I don't really care what laws are there.
It means that hosting public data in a US-based cloud means opening up yourself to a single point of failure!
I'm not entirely sure whether you can submit a DMCA takedown request if you're not a citizen of the USA. For once, you need to provide US-based contact info.
However, I am not sure if the service provider needs to check whether this contact info is correct before takedown.
So it seems that any public-facing data on any US-based server can be made to disappear for at least 10-14 days. But just because a country has crappy laws doesn't mean you should use those laws against them! ;)
> I'm not entirely sure whether you can submit a DMCA takedown request if you're not a citizen of the USA. For once, you need to provide US-based contact info.
So you can't just have an agent do the request on your behalf?
It doesn't. Our most recent copyright overhaul (which this is a feature of) currently works close-enough to OK to be tolerable to those who would otherwise fix it.
The basic idea is that for a provider (Github, Google, etc), to keep their Safe Harbour protections, they have to act like a 'dumb pipe' and just obey these notices. Just like how the phone companies trying to keep their Common Carrier status[1].
Even without the DMCA, if the service provider interjects itself into the dispute they risk legal action against themselves. For example, they could evaluate the claims wrong, and get sued to allowing actually infringing material to stay up.
It goes like this:
1. DMCA is filed by a person/entity with the service-provider.
2. Service-provider takes down offending content.
3. The person/entity that submitted the offending content now has the option to file a counter-notice with their service provider claiming that they feel/know there is no infringement. If they do file a counter-notice, then we continue on. If they don't file a counter-notice, then the story ends here.
4. The service-provider notifies the person/entity that filed the DMCA notice that a counter-notice was filed. The service-provider can now restore the content, but there is some weird stipulation that they need to wait something like 10 ~ 14 days before doing so.
5. The person/entity that filed the original notice now has to take legal action against the person/entity that they feel has infringed their copyrights if they want anything more to happen.
The party that files the DMCA notice stipulates that they believe that their work has been infringed under penalty of perjury.
[1] For a long time, people on the Internet believed that 'Common Carrier' provisions applied to ISPs, but it really only applies to phone companies. The Safe Harbour provisions of the DMCA are meant to give something similar for ISPs, though there are different rules.
1) what happens with perjury? how severe is "penalty of perjury"?
If at point 5, LayerVault chooses not to take legal action (say, because they realized their mistake), did they just take some random piece of work offline for 10-14 days without any consequence?
What if they do take legal action and lose?
2) You say the provider has to act like a "dumb pipe" and just obey these notices. Does this also exclude basic sanity checks such as calling back the phone number on the notice to check whether the corporation in question actually filed this notice, or whether notice-filer actually exists, whether the phone number is actually connected to the corporation the notice claims to be from, etc? Because otherwise, it's just a matter of time until somebody is going to ... take down all the things.
I just checked Wikipedia on perjury and it says, "Statements which entail an interpretation of fact are not perjury because people often draw inaccurate conclusions unwittingly, or make honest mistakes without the intent to deceive. Individuals may have honest but mistaken beliefs about certain facts, or their recollection may be inaccurate, or may have a different perception of what is the accurate way to state the truth", I think this is pretty much the case for LayerVault. So they won't get penalty of perjury, because they did believe their work had been infringed ... even when it's based on a wrong idea of what constitutes an actual infringement instead of a "heavily inspired by" rip-off?
| 1) what happens with perjury? how severe is "penalty
| of perjury"?
Penalty of perjury probably varies from jurisdiction to jurisdiction. I'm not a lawyer, so I'm not entirely sure. I would presume that this would be at the Federal level since the DMCA is a Federal law, as is copyright law. According to Wikipedia[1]:
> In the United States, for example, the general perjury
> statute under Federal law classifies perjury as a felony
> and provides for a prison sentence of up to five years.
So the person signing the DMCA takedown notice faces up to 5 years in prison and a felony conviction. That said, perjury convictions don't happen very often to my knowledge because they are so hard to prove beyond a reasonable doubt. Even if the claim looks like any normal person should have realized that it was bogus, it's possible for the claimant to attempt to say that he/she is dumber than the normal person.
| If at point 5, LayerVault chooses not to take legal
| action (say, because they realized their mistake),
| did they just take some random piece of work offline
| for 10-14 days without any consequence?
They opened themselves up to the possibility of perjury charges. The level of risk for them is debatable though. Other than that, there are no penalties to filing bogus DMCA notices. This is one of the many (valid) criticisms of the DMCA.
| What if they do take legal action and lose?
Then whatever the courts decide at that point. Taking legal action doesn't really have much to do with the DMCA when it gets to that point. This provision of the DMCA was all about taking things down quickly so that the claimant has time to get the court system moving (rather than needing to wait a few days to take something down, time during which they could be financially taking losses[2]).
Someone taking actual legal action is pretty costly. If someone uses a DMCA takedown notice, and follows up with a lawsuit, then this process is working as intended, even if the claimant/plaintiff is in the wrong. Who is wrong, and who is right will be handled by the courts at this point.
The real danger is using the DMCA without filing a lawsuit. There are few penalties[3] for false claims, little cost and yet a big benefit to sending the takedown notice. The service provider has to comply, unlike with a C&D letter.
[2] Think someone posting a product prior to release. E.g. posting Windows 9 a week before the release date.
[3] Perjury is a serious penalty, but you're not punished until you're dragged to court over it (and convicted). Thus far, I don't believe anyone has been convicted of perjury over a false DMCA notice, despite some really egregious examples (e.g. one of the takedown notices sent to MegaUpload was for a "url" that was actually a paragraph of text containing something like 90+ words -- pretty obviously generated by an automated process, but signed off by a human).
> did they just take some random piece of work offline for 10-14 days without any consequence?
As stated, in this case it's very unlikely there would be any consequences because perjury convictions are rare. However, it's worth noting that if any individual made a habit of this behaviour (false DMCA takedowns) then the probability of facing a charge would likely escalate.
Thanks for the full explanation. So are there no repercussions on the filing party unless the defending party seeks legal action? If the system is blind and you have a somewhat decent legal team you could surely hamstring your competition particularly on more subjective cases like this. 15 days can be a big hit (eg taking a site off google around Christmas)
Can you revoke a DMCA request you issue? This would be the reasonable way to handle mistakes. If the defending party, clears up all misunderstanding and proves to the other party that they made a mistake before 15 days is up, they should have to pull the DMCA request or suffer a larger burden of proof that they were acting in good faith.
Information obtained after the DMCA takedown request was issued shouldn't affect whether or not the request was made in good faith ex post facto. Currently there are no penalties (specified in the DMCA) for not withdrawing the request once it's become clear that it's bogus. There may be other ways under the law to get punished for a lack of action though. E.g. there may be case-law that states you can be penalized for a good faith action if you don't make motions to correct the mistake once you realize it.
Yeah, I understand how it is currently. I was saying it should (in a perfect world) be a much harder battle to prove you were acting in good faith if you are provided clear evidence that your request is completely wrong and you fail to redress the claim. The time period for a take down can cause a lot of harm to sites in exactly the same way that leaving up infringing stuff can.
Unless the infringement is real, where the harm to the original creator can be just as great. That waiting period gives them time to go to court and get an injunction before a falsely filed counternotice puts the illegal copies back online, potentially ruining a product launch or otherwise devastating their brand/income from the original.
Even if the infrigement is real the waiting period makes no sense. What difference are those 10 days going to make? Not to mention, the risk to the counter-noticer is quite high since they'd be commiting perjury if they filed a false counter notice.
It's totally skewed- a false counter notice has a high risk and a false notice a low risk. The damage due to a false counter notice is likely small (it can't disrupt the real owners own usage; it's probably been online quite a while anyhow; and the infriger(s) can find other ways to do so if he really wants to - notice the quick Flat UI resuscitation). But the damage due to a false notice is potentially quite large since it imples complete loss of service (not just someone else getting a few days of free reign), and potentially loss of momentum, and the issue with free speech - which isn't critical, perhaps, but all else being equal I'd prefer we err on the side of open.
So given the skewed potential for damage (much greater for a false notice), and the skewed risk to the fraudulent/mistaken party (much smaller for a false notice), it's utterly obvious there should not be such a waiting period.
The entire point of the provisions was a way to immediately take down the content pending things like injunctions, court orders, filing lawsuits. If we're talking about a site hosting a pirated copy of Windows, for example, the Microsoft could be losing money with every copy downloaded while it takes them a week or so for the proper judicial procedures to be followed to get it taken down.
The DMCA has good and bad parts about it. It's like a "Good Idea, Bad Idea"[1] short:
----
Good Idea: Make it relatively easy and cheap to send takedown notices so that the 'little guy' without fancy lawyers on retainer can participate.
Bad Idea: Make it relatively easy and cheap to send takedown notices so that the 'big guy' can send out millions of them relatively easy.
----
Good Idea: Make it so that you only need a 'good faith' belief to file a notice, so that the 'little guy' doesn't accidentally get caught up in legal terms and procedures.
Bad Idea: Make it so that you only need a 'good faith' belief to file a notice, so that the 'big guys' can get away with filing obviously stupid takedown notices because their fancy lawyers on retainer can easily quash any perjury charges over bogus takedown notices.
----
Obviously none of the 'little guy' stuff was considered during the drafting of the legislation though. It's just a by-product. Legislators also never considered the implications of bad actors, because they were only thinking of large content companies (which would obviously never do anything bad). It's also worth it to consider that this was passed on 1998 (and drafted earlier). The Internet was not as pervasive as it is now (even though the dotcom bubble was in its early stages).
Is there any checking that the named submitter actually submitted it? Anything to prevent someone forging a DMCA? Daft to do no doubt, but still an option to damage a company.
As posted a couple of days ago it seems like the only penalty of perjury matter is that you are the copyright holder or can legally represent the copyright holder.
I also don't see what would stop someone from sending thousands of DMCA takedowns from TOR or some anonymising VPN, laying waste to an entire website for 10 days.
You're also required to have a good faith belief that the content is actually infringing. Not easy to prove you're not, but it's something.
The EFF's "dancing baby" lawsuit is still going; if they win that will set a stronger precedent for forcing copyright owners to first consider fair use.
These kind of cases make me wonder, what ever happened to the common sense.. The (classical) Roman law was probably the best system the 'makers of law' ever came up to. After that, it only went downhill.
perhaps dmca takedown notices should stipulate that the party should leave some identifying information, so they can be traced back. I suspect credit card info would be a good one.
Basically. But if someone files a fraudulent or incorrect notice, they open themselves up to having to pay damages and legal fees to the injured party.
It was actually down at the time of that post too. I know because I was using Flat UI in a new project and had to ditch their vector illustrations which were the items in question from this DMCA.
I was not aware from the original post that the project had actually been taken down, and now thanks to pyre's comment I know a bit more about how the DMCA takedown process works. It seems to me that the process is quite vulnerable to Denial-of-service attacks, as demonstrated by the case in question.
They took it down, but you used to get a 404 when you visited the repo. This is the first time that it explicitly says that a DMCA request was executed.
Those that have used both are bigger fans of Pixelapse because it doesn't automatically save a "version" every time you save the file in Photoshop. Their payment tiers are also much more consumer-friendly: https://www.pixelapse.com/pricing vs. https://layervault.com/pricing
So I think the Noun Project should just send a takedown notice to LayerVault. Since the perjury clause only pertains to contact info and whether they can represent the project itself, that would have the same consequences for "accidentally" getting the actual allegation wrong (=none), right?
Sure, but if LV can argue it owns the rights to a similar-looking "ripoff", then another icon project can claim likewise in respect to them. Got a newspaper icon, or a stylized human emoting something? That's enough, apparently.
I'm not a lawyer, but it seems that the consensus here is that LV isn't allowed to assert that they own the rights, and therefore may be perjuring themselves.
You can do that, but you perjure yourself in the process, and won't take it down permanently. I could take down Rails by filing a bogus DCMA, but I'd have a lawsuit on my hands if Rails decided to pursue it
Did LayerVault try to contact Flat UI to discuss their grievances before sending a DMCA Takedown? If so, what was the result of that conversation. If not, why didn't LayerVault pursue that path before posting this to Github?
We in tech need to be good to each other because other corporate interests and patent trolls won't be. We should always give each other the benefit of the doubt and try to resolve issues without menacing legal language and the threat of lawyers. We need to show we can thrive without this garbage. Otherwise, it will only embolden our enemies and cause this type of confrontation to happen more frequently in the future - to everyone's detriment.
Layervault emailed the flat UI creator to ask him to take it down, flat UI refused because they didn't believe their work was infringing. Layervault kept pestering the guy, and eventually he gave up and removed a couple of the icons. And then layervault took this cooperation to be an admission of guilt and sent a DMCA takedown notice.
This all relates to trade dress, not copyright per se.
there's also the question of originality i.e:
Two men are wearing suits. Both suits originate in previous suits. The similarities that they therefore share would be those attributes of the suit that is not encompassed by previous works. So therefore, even though the two suits the two men may be wearing appear virtually identical, they are indeed not and do not share any substantial similarities to one another.
it's trés bothersome when designers behave like this when they're themselves just un/consciously influenced by previous artmovements and designers (florian freundt(2003), matias duarte spring to mind).
But then the question arises, is the person consciously aware of his influences or is he/she just building upon cultural memory? I mean were the DM designers at or about the moment where they released or designed their products had conscious exposure or borrowed either consciously or unconsciously using any of the imagery of LV.
There's a lot of unknown unknowns in this whole mess, well to be fair in all affairs of this sort, you'd have to know what the artist and designers were actually thinking when they were going through their creative process, which you nor I cannot do. For all we know, some of these items could be their inspiration.
We also don't know the reading habits or exposure to culture that they may or may not have had as young adults, or even as adults where they may have, for example, forgotten specific elements of exposure, and that's sort of one part.
the other part deals strictly with derivatives (creative meaning, not legal) What I mean is that strictly speaking, the new art movement influenced the "flat ui" movement, a folded newspaper is clearly an influence for the nuon project, dribbble iconry, DM flat ui representation and the LV icon. In much the same way a set of gears are derivative of a set of.. gears, which has been represented in too many ways in art that it's lost the entire concept of originality.
and Originality, due to Barthe's work on the Death of authorship, is quite moot at this point.
Oh, hey, I remember being angered by that story. That was a UK case and so hopefully it's irrelevant to the US. The US is IMO more reasonable than that with respect to parody and imitation.
LayerVault should have gotten a design patent. That works pretty well regarding trivial bullshit like corner radii, so it probably covers icon similarity also.
I have never seen the inside of LayerVault. The demo page only has vague outlines of similar controls. Does anyone have screenshots of LayerVault in action?
Those are pretty old pictures, and for sure it's not clunky. You can try out the 30-day trial for free. Nowadays it looks like this: http://cl.ly/image/1k230I2l421S
Funny that the mods haven't fixed the typo in this title in the 4 hours since it was posted, they must be too busy messing up other titles by removing useful context and deleting swear words...
If you want to speak up, don't hesitate to contact LayerVault's support ( support@layervault.com ) and let them know, respectfully, how you feel. You can also tweet your opinions @layervault.
I stress respectfully. Try to be well spoken and sincere.
I actually did. I was told by @Allan -- the founder of LV -- that I was "trolling", "writing inflammatory blog posts", that my "entire base of argument was based on the wrong facts". All in all, a very obtuse strawman argument, so being well spoken, sincere in many ways, will only lead to that sort of reaction from LV.
I disagree with this. When you have an issue with a company, that company is going to try to use all the little power it has to turn the issue in its favor. If you contact them, you're giving them power (the power not to answer your emails, the power to reply privately in a harsh way, etc.).
So go public. On a medium which the company doesn't control. And then they're not in control anymore.
It's exactly the same with SO: everytime someone comes up with a very valid criticism about SO there are SO officials (or high-rep users) saying "Put this on meta". But no, that's precisely the point: do certainly not put it on meta because once it's on meta it's the same little clique who's in control.
My opinion is that something like this reaching several times the front page of HN is the best way to make people aware of the problem (including the people at the company/ies concerned).
You can try a lawsuit for anything, about anything. It comes down to how likely it will be held up in a court of law, and generally, it helps if a precedent already exists.
As unlikely as it may sound, there have been lawsuits somewhat like what you describe. Perhaps most notorious is a tattoo artist claiming IP on Mike Tyson’s face tat visibly shown on Ed Helm’s character’s face, being used without permission. The artist sued Warner Bros. (I believe) for damages and asking to be taken down. IIRC the judge was very close to issuing an injunction against /The Hangover 2/. Not sure what ended up happening; probably settled out of court.
I hope that won't set a precedent on Github and we won't start seeing abusive and/or automated dmca's on repositories all around because "similar images were matched", youtube-style.
It's hard not to lose a little respect fo the LayerVault guys, even if the icons in question were direct copies. I get you're protective of your art, but this doesn't really make sense from a business standpoint, and their response has been less than convincing. It's really disappointing.