Vidly (YC S08) first to launch HD video on Twitter
techcrunch.com
techcrunch.com
x264 - core 57 - H.264/MPEG-4 AVC codec - Copyleft 2005 - http://www.videolan.or
g/x264.html - options: cabac=0 ref=3 deblock=1:1:1 analyse=0x3:0x113 me=hex subm
e=5 brdo=0 mixed_ref=1 me_range=16 chroma_me=1 trellis=0 8x8dct=1 cqm=0 deadzone
=21,11 chroma_qp_offset=0 threads=12 nr=0 decimate=0 mbaff=0 bframes=2 b_pyramid
=0 b_adapt=1 b_bias=0 direct=3 wpredb=0 bime=0 keyint=60 keyint_min=25 scenecut=
40(pre) rc=abr bitrate=2000 ratetol=1.0 rceq='blurCplx^(1-qComp)' qcomp=0.60 qpm
in=10 qpmax=51 qpstep=4 ip_ratio=1.40 pb_ratio=1.30
Thanks!
If you need any assistance, drop by #x264 on Freenode.
I'm always surprised by how video startups--whose most important technical resource is their video conversion system--don't seem to do sufficient research into how it actually works (or for that matter, how well it actually works). This may be because of the assumption expressed in posts like http://news.ycombinator.net/item?id=933713: that all software is pretty much equal and it doesn't matter what you use or how you use it.
Then again, I can't be too harsh, because you at least used open source tools instead of Flash Media Encoder, so you must be doing something right ;)
Thanks!
Also, 67-68 isn't particularly recent; the latest "stable" would probably be r1318 aka fe83a906ee aka Core 78. You can try trunk as well, but we're having some miscompilation issues with the latest improvements, so I wouldn't trust it for production use yet.
Another note is to check your settings; a lot of people have all sorts of weird encoding commandlines (you included, given the contents of the string I pasted above) from the days when x264's setting system was baroque and confusing. Now the defaults are good and there's some nice automatic presets that let you easily pick a tradeoff of time vs. quality (see x264 --help), so you don't have to mess with weird options.
This seems like an issue with a great deal of nuance, and I'd find such a resource invaluable.
I'd recommend Staxrip, Ripbot264, or Handbrake; all are relatively simple GUIs built around x264. For Handbrake, make sure you get the latest snapshot (the last release, 0.9.3, is very outdated).
On Linux or Mac, there are fewer options, but Handbrake is still one of them. If you prefer a commandline interface, see http://rob.opendot.cl/index.php/useful-stuff/ffmpeg-x264-enc... .
Handbrake has a great support channel on Freenode (#handbrake) as well.
See:
http://x264dev.multimedia.cx/wp-content/uploads/2009/08/qual...
http://x264dev.multimedia.cx/?p=164
http://x264dev.multimedia.cx/?p=102
As you can see, a bad H.264 encoder (e.g. Apple's) can barely compete with Xvid.
The entirety of my work on x264 is contained in the past 2 years of development. The psychovisual optimizations alone probably account for at least a 50-70% improvement in compression quality.
Great example of why I love HN (and IRC).
Option 1:
1) Open youtube.com.
2) Create a youtube account (First time only).
3) Post video to Youtube.
4) Open twitter.com, or a Twitter client.
5) Post a tweet with the Youtube link.
Option 2:
1) Open vidly.com (or client?).
2) Post video and tweet to Twitter.
Actually YouTube integrates with both Facebook and Twitter now. Go to My Account > Settings > Sharing and you can link your Facebook and Twitter accounts.
Once you upload a video, it will automatically update your status for both services.
I use YouTube for my uploads mainly because it's one of the built-in upload options for Android. I believe this is true for the IPhone too.
One of our primary goals is to remove friction. Reduce the number of steps it takes to distribute video to the audiences that matter to you. The ideal number of steps: 1. We're almost there.
Convenience is another goal. People are lazy, I am lazy, and I prefer something that is convenient and does all of the work for me.
Tell your parents to go to Accounts > Settings > etc, and watch their eyes glaze over.
Apparently, youtube can already tweet the videos automatically that you uploaded. What will be the shelf-life of that startup then?
Again, humbly sounds to me like a great example of a feature, not a product.
Not sure if you are the one who downvoted me.
I'm not saying YouTube's integration is better. I was just addressing the steps specified by crescendo on Option 1.
Don't think it was a slight on Vidly.
It only took as long as uploading the video. Maybe typing in a name. If you wanted to make it take longer, you could spend another 30 seconds geo-tagging it.
I did this about every day or two. Is that enough experience?
Biggest problem I had was intermittent and dropped internet connections. But the new Vimeo desktop tool is great at handling that.
I think you just wanted to say something insulting about Vidly, but couched in a form that seemed intellectually respectable. In content your comment is isomorphic to the kind of troll posts you see in TechCrunch threads.
I simply can't fathom how this can end in a good way:
startup = {name:"twitter"};
while(true) {
if (startup.isHyped()) {
startup.getMoreFunding(); // People invest in things that have hype
// Others see it as good idea to build something based on it!
var newStartup = {feature:["search","share video","share numbers","twitter porn","twitter lolcats"]};
// newStartup implements the feature
// Child hypes up the parent some more
startup.hype += newStartup.hype;
startup.acquire(newStartup); // When you're 100% reliant on the parent, where else can you go :/
}
}
I certainly wasn't trying to insult Vidly, more try and understand the rationale for such a startup existing. As I say, if it's facebook, being able to share video is just an extra feature they add. Like 'search' and handling urls properly. Why so different for twitter? Why must there be a startup dedicated to searching tweets, that twitter acquire? Can't they search their own data? Why does another company need to do url shortening?
As an exit, surely if you build on something else, you really limit your chances of being acquired to that parent. And if the parent is just funded by investors and not revenue... :/ Just seems flakey to me.
Do you think vidly can be profitable? Isn't it pretty risky building on twitter? What if twitter blow up? Do you expect any other potential exit other than twitter acquiring them with their own funding?
So sure, Vidly could IPO. Their path from here to there would be as tortuous as Microsoft's, but all the best companies are transformed as they grow. All you need initially is a starting point, to engage with users.
You are in effect framing the question as "could Vidly, without writing another line of code, have a successful outcome?" Probably not, but that's not the question that matters. The question that matters is "could Vidly, starting from this point, have a successful outcome?" The answer to that is, of course they could; any point where you've made something people want is probably good enough, barring certain pathological exceptions.
When you and the TechCrunch trolls make fun of startups launching their initial version, you're like someone making fun of Dell when it was just Michael Dell assembling computers, or Microsoft when it was just a pair of undergrads writing a Basic interpreter. How could such companies possibly grow huge?
Furthermore, the simplicity of what they do will be cloned and aped in months if not weeks, leaving them scrambling to add value, thus complicating their offering, effectively killing what little value they have now. The other problem here is that there are heavy brands that already exist that can sink the whole ship if they chose to, creating a mountain of Everest proportions for Vidly to climb.
Look at how many URL shortening services exist today. Simple ideas get cloned quickly, then eventually open sourced, shorting the original business to offering infrastructure over product. Unfortunately the infrastructure for Vidly is dumb simple. Give me an account an encoding.com and twitter library and what more do I really need?
And before anyone goes off on the famous I could build that in a week, keep in mind I built massify.com's self-scaling cloud-based HD video encoding solution in about a month, so I know what I'm talking about.
Why don't you cut these guys a break? They've just launched. When I recently wrote
If your first version is so impressive that trolls
don't make fun of it, you waited too long to launch.
Micro-Soft solved problems that nobody else was solving. And the problems they were solving where enormous. Vidly solves a small problem in a very narrow market. The two just don't add up, regardless of how broadly you want to paint it.
I'm curious if we're trolls because we have an opinion, or because we don't share yours? What would be the point of this community if nobody had an opinion? And regardless of the means by which these opinions are phrased, you might want to consider that a community of startup focused people whom are having an unfavorable reaction to a business idea really means. If HN is trolling an idea doesn't mean we are becoming techcrunch, it might mean there is something fundamental about the business idea that isn't jiving.
The distinctive quality of TechCrunch comment trolls is that they shit all over an idea that is just the first thing launched by a company that will morph dramatically in time. They act as if what the company launches with is the only thing it will ever do.
The worst you can honestly say about the initial launch of a startup using the launch-fast-and-iterate model is that this initial product doesn't appeal to you. To pursue the company itself as eagerly as you have done shows malice.
Are we in the same thread? The feedback was on the original idea. Because that's the only information we have on the company/founders.
It's fine to provide constructive criticism, but these threads are not constructive. They're wholesale character assassination and haughty dismissal of entire business strategies.
This is a "your baby is ugly" situation. Sometimes the baby is ugly. Don't mean it won't turn out to cure cancer, don't mean it ain't a great thing, just it ain't hitting on much with us so far.
In these situations quite naturally the parents have tendency to be overly-touchy. Internet posters, and this group especially, have a tendency to be arrogant jerks (myself included). I'll just say it: some folks need to get a thicker skin. Some others learn to be more democratic. Lots of blame to go around everywhere.
Internet text, whether posting, email, tweets, or whatever, is just really tough to do well.
But for this board to work like it is supposed to -- providing help and assistance to startups -- it's critical to be able to provide negative feedback, even if we screw it up from time to time.
I'm sure they'll likely iterate on the idea, but I think feedback is always useful. I really don't see how you can claim it's "trolling" to give feedback on an idea. That's what makes HN useful IMHO.
If we want "That's nice dear, what a great idea" we go to our parents... surely?
If there's a sign of a bubble, it's companies building
things that already exist, on top of companies that
have no business model.
* I think this idea is risky (building on another startup with no clear business model)
* I don't see the value add (already exists)
I expected a reply saying why it's not as risky as it seems, and how they're doing something fundamentally different from all the other video sharing sites out there.
Sometimes we all take shortcuts when giving feedback. Sorry if it was taken as a snipe, insulting or a trollish comment.
In other words, get the "Review my startup" thread going a lot earlier than Demo Day.
Maybe these threads could be invite-only, or accessible only to readers over a certain karma threshold?
(Another epic stretch: from a community-management standpoint, this may also help bandage the "I got rejected"/"I got accepted" divide.)
Microsoft and Dell had a business model. They sold stuff.
Vidly also have a number of free competitors, Microsoft did not at the time.
Maybe they have far bigger plans than just "HD video for twitter". If so, it'd be good to know, and if they turn out to be cool, then great - I'll be first to congratulate them :)
>> "you're like someone making fun of Dell when it was just Michael Dell assembling computers"
Again, the comparison isn't good IMHO. People pay money to purchase computers. There's an obvious business model there.
People currently don't pay to upload videos. The potential business model is far harder to see. Couple that with making it "for twitter!" and for me, it looks like a bubble - startups funded to make features to be acquired by another startup. Endless cycle of investor money being pumped in until something explodes.
Again, FWIW I wasn't making fun of. But IMHO vidly should rise above the "We're X for twitter". Hopefully they will.
-Twitter doesn't have a business model. OK. So twitter has a better chance of going out of business, maybe, then a company with a business model. Are you saying there is no chance twitter will be around in 5 years time? No? then what is the point of that objection. Who care what Twitter's business model is. You only care about it inasmuch as it affects the stability of Twitter as a platform.
-Vidly has free competitors Are you applying this as a general rule? Anyone in a type of business where there are free competitors is disqualified?
Sure there are obstacles. The chances of any little company of getting massive are tiny almost by definition. But you haven't put forward anything disqualifying, which is what you are implying.
Here are some possible rationalisations for such companies existing, such companies betting on acquisition by a limited number of buyers & other concerns:
- Smaller companies can sometimes do things better.
- Smaller companies can sometimes do things cheaper.
- Small companies can fail, so they can try riskier things.
- Small companies may not have as much of a brand to protect, so they can try riskier things.
- More features can be put to market test by multiple indipendant companies then by the 'parent' alone. Individual features can be put to market test and good ones have a better chance of being discovered.
- Features can be tried multiple times by multiple companies, with the best one ultimately emerging.
- Startups (using a certain definition) have a 1/x chance of succeeding at launch. Success/failure is binary. The risk of faling completely is extremely high. Therefore, a 10% or 20 or 40% chance that Twitter will disappear while they rely on it to survive is not as scary as it is for an established company. If your chances of success are only 1/5 anyway, going to 1/6 is not too bad. Basically, startups are structured to handle that sort of risk.
These reasons may or may not add up to a net efficiency gain overall, better features or whatever. Redefining what companies do themselves vs what they outsource is an interesting part of how economies evolve. Outsourcing innovation (even if it is just a small feature) might be a good move, sometimes. The market gets to decide.
In contrast, Facebook has over 10x the people, and 10x the user base, but does major UI shakeups regularly, tries out new features often, and isn't afraid to tweak or remove them if they don't work out. They also are doing interesting things technically (e.g, Cassandra, Scribe) and open sourcing them, some of which Twitter deploys themselves (e.g., well, Cassandra, Scribe).
Facebook definitely seems to fit the startup ideal much more than Twitter, yet Twitter is much smaller. What gives?
Twitter is extremely startup-like in one critical respect: their attention to how the community is using their product. That's how they became huge. It's the fact that they became huge by focusing on this while others who happened upon the same space did not do this, and thus did not become huge, that wins them respect in the startup realm. Much of what you call their risk-averseness comes from this desire to follow rather than force their users. It's a lesson many of us would do well to learn and, I suspect, one of those things that is far deeper than appears on first sight.
It'll be interesting to see how well that works when/if they try to monetize those users.
Twitter is entirely unproven. It's a big mystery what Twitter's actual user engagement numbers are, so nobody really knows how sticky the site is. How do you know Twitter's approach is working? It seems entirely too premature to say.
Honestly to me it seems like either Twitter's user engagement numbers are really bad, which is why Twitter doesn't talk about them, or Twitter's analytics are so poor that they don't know how their users behave, which doesn't really reflect well on them either.
In that light, it makes sense to me that Facebook can afford to take more risks that twitter. And Microsoft can afford to take more risks than Facebook, etc.
That contradicts a lot of sentiment on HN.
> And Microsoft can afford to take more risks than Facebook, etc.
Where is Microsoft making more risks than Facebook? They are quite conservative about the cash cows. Windows has a lot of complexity to maintain backwards compatibility going back decades, and that complexity has a very high cost when it comes to making a stable and secure product.
I define risk as the unknown part of the equation. In that sense, startups (particularly the customer development types) are in the business of flushing out risk through validation before action.
Where is Microsoft making more risks than Facebook?
Zune, Bing, Maps, whatever that table-top display technology is, etc etc. These are all risks insofar as they are outside of MSFT's original core competency, which I understand to be operating systems.
They are quite conservative about the cash cows.
Facebook isn't conservative with their cash cow? Are they revolutionizing online advertising somewhere that I'm not looking? All I can see is targeted banners. Looks very conservative to me.
First, small is relative. If 'company' is 3 guys working on a project for a couple of years, maybe they can be bought out at a price that Twitter can afford.
Second, twitter is a big deal. That means it probably has the potential to be a big company. You may disagree and that is fine. But other people consider Twitter big.
If a company came into being specifically to be Twitter's accountants or lawyers, it wouldn't be that big a deal. You might thinks it is risky or narrow, but you wouldn't be calling it a bubble.
If you really think these are just companies manufactured for sale to Twitter (not self evident, but granted for the sake of argument), you can think of it as just a variation of that.
Hi. Good to see the CEO of the company posting on this thread. Just in case you were not aware, Twitter launched before you were able to tag other people in your Facebook status. Tagging a person, however, is just one of many differences between Facebook status and Twitter. There are no hash tags, no lists, and no trending topics on Facebook. More importantly, on Twitter you can get direct access to the words of celebrities and other notable people. On Facebook typically all you would see would be an HR maintained fan page.
There's a big difference between that and getting a direct line via Twitter. I'll give a concrete example. A couple of days ago I was able to ask Jason of 37signals whether he knew of a copy of his talk from Start Up School besides Justin.TV. He responded to me within an hour. Interaction like this is not possible via Facebook status updates. It seems to me that your claim that Twitter built something that already existed isn't true.
As a side note, your tone seems a bit confrontational for what I would have expected from a YC CEO.
My tone, which obviously text does not convey, was playful and light-hearted.
We all know you don't post news to YC to get fluffy love comments, you post it to get 100 reasons why you suck. We don't take it personally.
Ah, my apologies then. :)
Please share the answer.
How are you different from any general video sharing site on the net?
Facebook consider "Able to share a video in status update" as a small add-on feature. Why when it's in the twitter world is it considered a good idea for a startup :/ Surely you can appreciate to some, this seems bizarre.
Most of these comments are very narrow in vision. Vidly has a long-term vision that only begins with Twitter. We'll be sure to keep the community posted as we continue to release new features and slowly reveal the direction of the company.
We appreciate the doubt, without it, we wouldn't be having any fun.
Someone please spell it out for me.
- browse your file system for the video (starts uploading as soon as you select it) - while it uploads, fill in the title, description and tags. The title could be your status update, the other fields are not necessary if all you want is to tweet the video. - click submit and wait a few minutes (varies depending on their load) until your video becomes available for viewing - copy the url and compose a tweet including it
It would be easy to create a mobile or web app to streamline this process using the Youtube API(e.g. type in your tweet, chose/record a video, submit and forget. We'll fill in the details and publish your tweet when the video is available). It's probably been done.
Vidly looks like yet another video site, only it attempts to be Twitter-friendly by incorporating that feature from the start.
Not only is this possible with a few clicks -- it's already possible with no clicks at all.
It's an entire company devoted to replacing one feature that already works fine on an immensely popular site.
Just set up AutoShare. When you upload a video, you get a tweet (and FB and others get notified as well if you've set it up to cascade)
Where was the value again?
http://www.youtube.com/watch?v=_nGsF0YR7NU (instructions on how to set this up on YouTube)
I'm doing something similar at the moment - a combination of Twitter client for j2me phones that beside normal twitter functions allows taking photos/videos/audio recordings and a web site to display recorded material.
I've been live for about a month, and surprisingly I'm not seeing many users recording videos. Most would use it as a normal Twitter client, and some maybe will snap a few photos.