Genius laid off a bunch of its engineers
theverge.com
theverge.com
The only place I've personally seen it used is for people to snarkily mark-up political op-eds.
It's a cool vision -- I wish I could see, for a given article, the best inline comments or just those of people I find credible -- but it doesn't seem scalable to the wide web, as there just aren't enough people willing to put smart commentary on every article instead of just firing off a few tweets, commenting on the article at an aggregator like reddit (which gives a signal of which things are worth commenting on, and a built-in audience for the commentary) or writing their own post.
"Annotating other people's content" is one of those evergreen ideas that keeps popping up (here's an article I wrote about a startup doing the same thing, way back in 1999: https://jasonlefkowitz.net/2006/12/third_voice_wro/), but every time someone tries it implementation issues seem to kill them.
https://www.youtube.com/watch?v=U1swp-2QeMI
We worked as an embeddable product like Genius as well as a browser extension, and we had similar issues: client-side performance, "how do you anchor things to a page that's probably going to change out from under you," and "how do you scale to serve the combined traffic of every site you serve comments on without spending more money than you make?"
I think web annotation is a good idea but probably needs to live in a moneymaking niche for a while before it can grow to the broader web. Lyrics were an interesting start for Genius but it's tough to make money there. Law, news, internal research, etc. seem like interesting starts to me.
To add to your point, Disqus is also struggling: https://techcrunch.com/2016/12/13/disqus-lays-off-11-as-it-p...
I mean, seriously:
But more recently Disqus has been trying out other things to expand its business. That has included a growing programmatic ad platform based around sponsored comments, which debuted in 2014 (it’s this that makes the company money, as Disqus takes a share on ads that run through its otherwise free platform). And earlier this year, Disqus opted for another traffic gain as it relaunched of its own site as a central repository not just of content and comments from sites that you follow and comment on with a Disqus ID; but also of standalone comment threads that existed independent of these, Reddit-style.
This seems like a lot of digging of one's own grave. Apologies if all of this ground was covered in Dec, but it looks like the story never made it to HN.
What's interesting is that Genius seems to have run into the same brick wall Third Voice did: people who make Web sites really hate having randos mark them up. See, for instance, these two articles on Genius from last year:
http://observer.com/2016/03/genius-web-annotator-emma-dawson...
http://www.slate.com/articles/double_x/doublex/2016/03/news_...
The complaints in those pieces are exactly the same ones that people were making about Third Voice back in the '90s. Exactly the same. And Genius' defense ("But your blog is public!") is exactly the same too.
I feel like maybe there's just a fundamental issue of empathy here: the kind of person who launches an annotation startup can't understand the complicated emotions the kind of person who writes in public has about their writing. So they keep stepping on toes by barging in too strongly, instead of looping the content creators in early to get them on board (or at least figure out how to make an annotation product that doesn't offend their sensibilities too much).
And for news specifically, there are tools like Fiskkit [1], that offer tags that are used a lot in that context: logical fallacy, untrue/uncited statement, etc. Perhaps these domain-specific tools are the only things that will ever catch on (like Rap Genius originally did).
I get the annotation idea, but how the hell do you make significant money off of it?
2. Have a somewhat successful user generated content site.
3. Mutter some generalities about how we're "not just X man, we're like totally X for the entire internet"
4. ??
5. Profit
I found their "right kind" to be cringe worthy:
"While Facebook’s mission may be to connect the world, Rap Genius' mission is to bring context and knowledge to those connections. Rap Genius is itself a kind of social network, but its a social network based around texts and discussions of texts."
[0] - https://genius.com/Marc-andreessen-why-andreessen-horowitz-i...
Still don't understand how or what they showed about how are they going to make money, I do understand the "great vision".
did we have a bubble with no burst? just a fizzle?
"This is what a lyrics site looks like now. You go, you get a big ringtone ad. This is terrible! Where can you find the lyrics?" [Lehman] then mocks the site, as the crowd laughs at the absurdity of an ad for breakfast cereal next to the lyrics of Lil' Wayne...
"We're basically going to dominate the lyrics space. We're going to kill MetroLyrics, kill all these sites. After that, all texts. Poetry. The Bible. Literature. Tax code."
Strangely, the other areas are mentioned in a blur, with no mention of what will actually be next after rap lyrics."
It is integrated with Spotify although it seems to pop-up randomly for me. I've never figured out why some songs show info from Genius or how I'd get to it intentionally if I wanted to (but I admit I haven't really tried).
An example: https://genius.com/Barack-obama-state-of-the-union-2016-anno...
I like the idea of annotation and I even sort of like Genius's implementation of annotation. There's a pretty decent annotation of Gatsby on the site. Virtually every line of The Waste Land has a good annotation there too. But, not to put too fine a point on it, this stuff all seems like, from a platform perspective, something Dustin Curtis could have built solo.
It's just that annotations, by themselves, fail to do that. Their VCs must've been betting either on some new killer feature that would come out, or that users would react to the existing annotation product differently than they did. For a VC, though, "could rival Facebook" is often more important than "has a small chance of doing so".
And that annotations could eventually be the successor to eg Disqus across the web, as well a new type of social network.
I think the world's information would be so much richer if it had played out. I'm still bullish on annotation tools for empowering knowledge.
But the annotations community outside of rap songs didn't take off in the same way.
Their only other major play is annotating things like scientific papers but I cant see the market for that being big enough to make money.
99% of lyrics sites are utter trash littered with popups, redirects and extremely obnoxious ads. Genius was the first lyrics site I ever saw that broke that mold.
Sure, they also broke Google's arbitrary rules, but they had a better product that genuinely deserved to be at the top of the rankings. I still wince when I click on a link to any other lyrics site.
The lyrics licensing industry is absolutely brutal, and I will be pleasantly surprised if anyone that isn't owned by a music publisher is able to succeed in this market.
How on earth were the rules arbitrary? These guys were exploiting and abusing backlinks to boost their page rank.
Not sure if this W3C spec[1] is in the right direction, but at least it's an open standard.
Personally, I liked the WebMention[2] model better (rel="webmention"), where you mark up HTML. But it would require someone to take up the banner and make it easy to distribute and deploy.
Annotation in a weird way is even harder than a "boil the ocean" problem; you have to have the exact right number of participants, across all the enormous complexity of the Internet, or at least, a very huge chunk of it. If you have nobody on some page, the annotation system is useless, and if you have thousands of people pounding away at the same page, it's pretty useless too. You have to hit this window of reliably having just the right number of users on a page, on an Internet that intrinsically tends towards power law page viewing, somehow across the entire Internet. After ~18 years of watching the space, I simply believe the entire vision is unsolvably flawed, especially in the presence of successful link aggregation and commentary sites.
The power-law distribution of web pages wreaks havoc on almost any attempt to distribute people well here. Your solution basically would make it so the annotation works OK on only those pages that are on the high end of the local power law distribution, but that is probably fewer pages than you might think as you'd all have different long tails.
The problem isn't that any given aspect of the annotation problem is unsolvable. The problem is that you can't solve them all at once to build a successful system.
Recall that I said I've been "watching this space" for 18 years now. That's no exaggeration; I was the first to hack Third Voice back in the day. It's not like people haven't been trying. Speculating about how it might work someday is really kind 10 or so years too late. In 1999, theorizing about how cool annotations might be was adequate. In 2017, a theory about how cool annotations might be needs to not just explain how they might be cool, but why all previous efforts have failed, and why yours might succeed. That's a much higher bar to leap now.
And one obvious solution is to trim away bits of the problem, but you end up in the local maximum of "link aggregation site with comments" pretty darned fast.
https://www.math.utah.edu/~beebe/support/html/Docs/group-ann...
Around the same time they were pitching colleges for classroom use, but were embarrassingly unprepared to address the copyright problem. Their "platform" involved republishing texts to which they had no rights. They did not even have a good story about claims on student (author) contributions. Who owns the annotations? They had no coherent answer during a demo at my uni. We passed.
It is beyond me as to how smart investors threw money at such incompetence.
1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes; (2) the nature of the copyrighted work; (3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and (4) the effect of the use upon the potential market for or value of the copyrighted work. (https://www.copyright.gov/title17/92chap1.html#107)
Rap genius was/is a for profit corp and they reproduced the copyrighted work in full, on their own servers. I don't see how they could argue fair use. In addition, their use is arguably non transformative, in that they leave the original intact.
The writing was on the wall, Sony deal or not. It was just a question of copyright holders noticing/acting on the violation. The initial round with Anderssen Horowitz for 15 million in 2014 prompted swift response.
I don't know if this is sufficient grounds to argue fair use, but it is clear that the case is not cut-and-dry.
The gold standard for fair use is academic citation. Genius use of the lyrics was clearly commercial (because it benefited financially and is a for profit organization). The terms "direct" or "indirect" are not in the language of the guidelines. In (4), it would be enough for copyright holders to show potential for future harm.
A strong case of fair use needs to clear each of the four factors. For example, just posting the lyrics online (without profit) violates (2) in that it is not transformative, (3) in that the work is reproduced in its entirety, and (4) in that it harms the "existing or future" market for the product.
By contrast, an academic citation, a sentence of a novel, for example, is non-commercial, transformative, partial, and non-damaging. Incidentally, things get more complicated with poetry---scholars working on short-form lyrical poetry often need to clear the rights for publishing their research. Poems are short, it is therefore difficult to avoid reproducing them in full, which violates (3). Courts prefer partial reproduction for fair use.
"Purpose and character of the use, including whether the use is of a commercial nature or is for nonprofit educational purposes: Courts look at how the party claiming fair use is using the copyrighted work, and are more likely to find that nonprofit educational and noncommercial uses are fair. "
Kind of a bummer that they raised at such a high valuation and repeatedly set their sights on scaling to areas with minimal chance of success.
Also always thought about how the need for annotations is probably a lot lower in other genres / types of text content. I rarely need an abridged Wikipedia-level explanation of who Donald Trump is, but these kinds of clues tend to be key in interpreting the wordplay of hip hop.
They're frankly turning into a hip hop publication at this point, much to my dismay. I don't find their content very attractive beyond their core competency of verified artists.
I never found most hip-hop lyrics being so deep that hey need some annotations for explanation.
Then it felt like very far fetched to me that people who considered annotating hip-hop lyrics now want to "annotate the world".
And now they can't live up to their promises, which perfectly fits in my, probably wrong, model of hip-hop culture in general.
For example I really like grime but I haven't lived in England for over a decade so I have no clue what a lot of their references mean when I first hear them because I never hear them in my day to day. Genius helps with that and a lot of other cases.
Jiggy-jabberjaw vitamin idol and primal rages
When bible page verse tidal waves begun (Water won)
I wrote the book that shook America to splinters
In a winter wasteland icicle-bound barren township
Call Baron Münchhausen and proud gimps to clinch this chemical war pig
Delinquent sinking like a paper tugboat well beneath your holidays
Beached on a red tide infested shore corroding eastern seaboard beauty
I'm a jigsaw slab, but all the pig saw was dirt bag
(https://hypothes.is/blog/annotation-is-now-a-web-standard/)
Also, if you're an LP of a16z, don't let Horowitz waste $40m just because he wants to hang out with rappers.
It's going to impact their bargaining position for a new gig
Apparently Genius didn't think so.
I think the article hinted at that.
I interviewed at Genius and started following her because of "Blondes make better engineers" tagline, it cracked me up.
Maybe they could donate the lyrics metadata to the public domain.
The annotations that journalists do on speeches can be super useful.
But in general... I'd like to have a portable system to annotate PDFs, textfiles, etc. Local files. Take notes and cross-link against multiple files & multiple notes.
I've been making notes sporadically for a decade about such a thing, with occasional bits of code that don't really solve the problem. Currently I'm playing with a neo4j backed proto-prototype. :: shrug ::
Especially for more symbol-liberal artists like Bowie.
And rap like Norf Norf.
The reason I'm complaining about this isn't to be a wet blanket- I just can't believe this service got the amount of funding it did. If their musical annotations are so cringeworthy and poor, what made investors think they could annotate other websites so well that a $58.9 million dollar investment would produce any reasonable return?
Genius reminds me quite a lot of last.fm in terms of what kind of service it was and how well it has fared as time has progressed. Both were reasonable ideas to begin with, but is there any reason these ideas needed millions of dollars in funding to operate? I am really getting the impression that these tech VC's are just out of ideas and are consistently trying to monetize internet toys which in reality should remain small web services for dedicated user-developer types.
What?! Up until then it had been speculation. This is one of my favorite features.
>doesn't provide any more value in its annotations than does azlyrics, youtube comments
AZLyrics are almost always a direct copy of Genius, errors and all; and youtube comments?!
>A lot of the time, its annotations are just straight up wrong
so you can (eventually) edit them! What other site lets you do that?
I don't understand how they make money but their site has a lot of value for me.
If I'm looking for song lyrics, and not some collaborative social information-exchange medium, then that's exactly what I want it to look like (and pretty much be.)
Genius is slow loading and slow scrolling and has bigger ad blocks than most lyrics sites. If I'm not specifically looking for annotations rather than mere lyrics, it's strictly worse Tha the competition.
> I haven't found Genius to be significantly better at annotating lyrics than any of it's competitors
So again, what competitors are they talking about?
Edit: Since I've been downvoted at least once, I want to clarify. I'm unsure what monetary value can be extracted from the ability to annotate third party websites, or in this case, specifically, lyrics. As others have mentioned, if I want lyrics, I typically just want the lyrics, as uncomplicated as possible. Perhaps I'm missing something.
With regards to annotation specifically, songfacts and lyricsmode are competitors in that they support the same kind of annotation Genius does. In this sense, they are certainly competitors, though Genius has far more annotations and is winning the competition.
My main point about the failure of Genius is that it's Jedi mind tricked people into thinking annotation, and not just annotation in general but Genius' specific implementation of it, is the 'best' way to discover the 'objective' meaning of individual songs. There is no objective meaning and Genius' annotation feature is hardly as powerful as people give it credit for. Genius is just a discussion board with annotations enabled- another iteration of the many discussion boards which VC's have foolishly pumped tons of money into and not received a payout from.
Why does that not make them competitors? With the struggles Genius is having I would say they are definitely feeling the competition.
Nobody else has anything like verified artist annotations.
I would love to see you Google or Reddit all definitions individually.. seriously.
Edit: yes, I do correct people's mistakes
(And my linked comment did contain a juxtaposition about a16z's purported investment strategy and their Genius investment, so I don't think the retroactive chiding is entirely called for.)
When a thing happens 99% of the time, its happening again doesn't mean there was any truth, or even any information, in your reasoning. You might just as easily have linked it to A16Z's cafeteria as their consumer strategy. Whether or not they have a cafeteria.
(I'm not assuming this startup has failed, of course; the point about snarky dismissals holds either way.)
I could actually see SV being an impediment to innovation by "salting the fields" in so many product categories with hugely overfunded companies that will eventually flame out.
The problem is that, as far as I know, YouTube has not been wholly profitable for Google at this point, so they've really been trying to squeeze all the money they can out of it, and I feel it's come at the expense of the service itself. I spend far less time on YouTube these days.
I would be much happier spending money directly on a service which treated its users as customers rather than the product.
As far as other features, I think it's naive to say that YouTube has every feature someone could want a video hosting service to have. I feel YouTube could improve quite a bit in plenty of areas, particularly in how people 'watch' music there.