A VC - Latest Comments in That's Only Ten Lines Of Code

Re: That's Only Ten Lines Of Code

Vivek Krishna — Thu, 11 Jun 2009 00:57:02 -0000

The point though is that , on sites like delicious and twitter there are network effects. Even if the implementation is easy ,the actual value lies in the user base and not the code because the value of the product is directly proportional to the user base. On a site like bit.ly which provides a simple url shortening service, how can a big user base be important?

Re: That's Only Ten Lines Of Code

Prokofy — Wed, 22 Apr 2009 22:20:39 -0000

I continue to be happy with bit.ly because it is one of the few services that doesn't stay annoying in your view for being a sharing device, but only will come up on the sidebar when you actually do something with it. It also keeps a record of how many people are clicking on your link, which is useful. And you can twitter right directly from its window after it shorts the URL.

I'd like to see these URL shorteners put up a daily list of URLs most shortened from which sources, like the NYT has "most emailed".

Re: That's Only Ten Lines Of Code

Matthew Mamet — Tue, 14 Apr 2009 15:58:51 -0000

The "that's only 10 lines of code" line has gotten so tired, that at PermissionTV we use it as the title of our developer screencasts. Our version is "5 lines of code (and a button)." http://www.permissiontv.com...

Re: That's Only Ten Lines Of Code

tweetip — Sun, 12 Apr 2009 23:30:33 -0000

Still in fucking awe of Twitter being twitter. We love bow hunters. We love all of you. You are what the tech industry could have been, should have been ; (

If you all cash out hugely tomorrow, Monday, April 13, 2009, everyone of you deserve it.

13x over :)

Re: That's Only Ten Lines Of Code

tweetip — Wed, 08 Apr 2009 01:23:20 -0000

Thanks Fred. Our career path offers almost no peer review, so your comments & support mean a bunch. The project I've been working on since '92 is important to Sylvi & I, and, if twitter stays independent, them too. I'm in awe of how Twitter is being twitter.

We've been scaling back here a bit - tired - I'm not sure you who pay attention understand the environment we have to create to analyze the data coming in. It's not all 'code' on a mac. I'm working on it even as I type this - - an EQ in LA is alarming my iphone.

This project is deep. It is more than 10 lines. It is NOT fun. If I succeed, it will change our understanding of evolution and reality.

If I don't succeed, I have already answered for myself what is real, and what is hard work :)

Re: That's Only Ten Lines Of Code

fredwilson — Sat, 04 Apr 2009 08:51:10 -0000

But its baked into twitter which is slowly but surely going mainstream

Re: That's Only Ten Lines Of Code

Winston — Sat, 04 Apr 2009 06:02:49 -0000

I don't think of it in terms of lines of code but TinyURL has always been one person's side project, and while the extra features of bit.ly are really cool, I'm not sure why they add so much overhead to what it takes to run TinyURL.

Re: That's Only Ten Lines Of Code

James C. Roberts IIi — Fri, 03 Apr 2009 02:50:08 -0000

Code length would only be important as a defense against copying and it would be folly, anyway as an infinite number of monkeys on an infinite number of typewriters . . .. You get the point.

Engaging the user (members of the "market") is the key . Here is my concern: Just how many people know what URL shortening is (other than a digital baking ingredient--OK, OK, bad joke)? It is not intuitive. Adoption may be a process limited to the technorati.

Re: That's Only Ten Lines Of Code

fredwilson — Wed, 01 Apr 2009 11:32:37 -0000

The same firm that started bit.ly is the only investor in tweetdeck

Re: That's Only Ten Lines Of Code

Malcolm — Wed, 01 Apr 2009 10:34:57 -0000

Hmm, before I research this some more, my first thoughts are:
1) with such a low barrier of entry for clones and super low user switching costs (significantly less than community-based sites like Twitter, Facebook, YouTube, etc).. the risk of losing your differentiating features is way too high
2) as a TweetDeck user, I'm curious how much this has helped bit.ly (and thus created a false sense of loyal users) since bit.ly's service is listed first in TweetDeck's list of many. I know I've used bit.ly many times without thinking about it simply because it was the default.

Re: That's Only Ten Lines Of Code

fredwilson — Wed, 01 Apr 2009 09:25:01 -0000

Excellent points Ian

This strategy has worked out well for our portfolio companies that have taken it, including tumblr, twitter, disqus, etc

But you are taking some risk with it

Re: That's Only Ten Lines Of Code

Nichelle Stephens — Wed, 01 Apr 2009 08:32:26 -0000

I stopped by the Bit.ly offices a few months ago, and when they show all the tracking features, I realized that bit.ly was more than a URL-shortener. It could possibly rival Delicious or ClipMarks.

Re: That's Only Ten Lines Of Code

Ian Betteridge — Wed, 01 Apr 2009 07:25:21 -0000

Well, the problem is that Fred's comment contains a contradiction. With no revenue stream, each users is an incremental cost to your business - in bandwidth, support costs, general server infrastructure, and so on. The cost per-user may be tiny, but once you build a large-enough user base, you're going to break that $50k a month limit.

And, if your business gains value to users via a network effect, the more people on it, the more attractive it is to new users. This can lead to your costs snowballing very rapidly. At that point, you either have to find a business model pretty fast, or sell more of your company to investors to fund it. And at the moment, venture capital - like all money - is in shorter supply than it used to be, which in turn means VCs will be able to demand more of your company than they perhaps could a year ago.

So, to my mind, the strategy of worrying about where the money comes from later is no longer one which has much to recommend it to a growth-focused business. I should stress that I don't know if Bit.ly has no business model - it may well have a clear, sound business plan to introduce paid-for services, say a "Bit.ly Premium" for corporate customers.

Re: That's Only Ten Lines Of Code

fredwilson — Tue, 31 Mar 2009 18:10:27 -0000

I am not sure about the "leave instantly when something better comes along"

I didn't see that happen to skype, youtube, facebook, etc, etc

Re: That's Only Ten Lines Of Code

fredwilson — Tue, 31 Mar 2009 18:10:14 -0000

And I referenced your views in the post. they are certainly one of the reasons we sat this one out, but not the only one

My point in bringing up delicious is that these url shorteners collect data about urls that people think are important, and like delicious, that data asset could be valuable

Re: That's Only Ten Lines Of Code

fredwilson — Tue, 31 Mar 2009 18:10:10 -0000

Joshua comments on AVC!

Re: That's Only Ten Lines Of Code

fredwilson — Tue, 31 Mar 2009 17:59:08 -0000

This is great

Re: That's Only Ten Lines Of Code

Ade — Tue, 31 Mar 2009 17:41:42 -0000

"barrier to entry" - This is the Warren Buffett theory and he has a point. $2 million to make url shortening? Personally i think it needs a better name, 'urli' up above was catchier

Re: That's Only Ten Lines Of Code

Peter Bowen — Tue, 31 Mar 2009 16:27:56 -0000

Ideas float in the ether... take one down, try it on for a while, and if you can't give it a go, it will just float back up into the ether. I remember a childrens' book by Roald Dahl called the BFG. There was a giant that would go around catching dreams and the distributing them to the children of the world. You couldn't really see them until they were in the jar. Some were good, and some were bad, but you couldn't always tell until it was in the jar. Ideas are like that.

The guys who expect millions of lines of elegant code miss the point. You don't build stuff and then sell it to people who you convince to want it... You figure out what is cool, you build it, and then you sell it to people who wanted it the whole time. If it takes 10 lines of code, then even better.

Finally, If you could have done it yourself - then why didn't you? Actually, why don't you... the market is open for guys who execute better. eBay, Google, Facebook, Compaq, and Enterprise Rent-a-Car were all "copycats" that did it better. My .02...

-Peter

Re: That's Only Ten Lines Of Code

Yule Heibel — Tue, 31 Mar 2009 16:12:00 -0000

Right! It also reminds me a bit of people claiming that they could execute drip paintings like Jackson Pollock did. Aside from the actual skill it takes to make those paintings, they probably wouldn't have that whole history of accrued acts and talents that comes together to advance a successful execution.

Re: That's Only Ten Lines Of Code

pwb — Tue, 31 Mar 2009 15:58:47 -0000

Amen. I'm so sick of that line. Then build it already!

Re: That's Only Ten Lines Of Code

projectshave — Tue, 31 Mar 2009 15:36:26 -0000

Where's the barrier to entry? If it's not tech, then it is your grip on the audience. But on the net, the early adopters are fickle and will leave instantly if something marginally better comes along. When the unwashed masses use your service, then you might have a sustainable audience.

Re: That's Only Ten Lines Of Code

Taylor Davidson — Tue, 31 Mar 2009 15:35:42 -0000

The funny thing is that users don't use bit.ly because of the underlying code.

Nicholas: spot on... the value comes from the execution, not the idea; users don't care about the ideas but simply how things work; but moreso, bit.ly's value is determined more by the robust product integration, the userbase and the data users create (and bit.ly manages) than the code itself.

Re: That's Only Ten Lines Of Code

Ivan Kirigin — Tue, 31 Mar 2009 15:21:49 -0000

Here is the code to do what bit.ly does in ~10 lines of code - using their API.
def makeBitlyLink( longUrl ): url = 'http://api.bit.ly/shorten?version=2.0.1&longUrl=%s&login=%s&apiKey=%s' % (longUrl, 'ivankirigin', 'redacted') request = urllib2.Request(url, None, {"User-Agent": "Tipjoy/1.0 +http://tipjoy.com"}) try: results = json.read( opener.open(request).read() ) if results['statusCode'] == 'OK': return results['results'][longUrl]['shortUrl'] except: pass return longUrl

Re: That's Only Ten Lines Of Code

joshua schachter — Tue, 31 Mar 2009 15:01:47 -0000

i like to think i had more consideration as to what's good for the ecosystem as i built delicious.