That's Only Ten Lines Of Code | Hacker News Reader

That's Only Ten Lines Of Code

avc.com

55 pointsph0rque16 comments

It is absolutely true that size of the codebase is not a consideration to business success. Mine would fit comfortably in the average CS101 lab project -- who cares.

The reason why bit.ly's triviality is a reason to worry about its future is not that it is short. It is that there is no defensible competitive moat -- they're primarily parasiting their users off another popular service (Twitter) who could obliterate them as a side effect of an engineering change which would be mostly unremarkable. (e.g. "Our front-end now auto-crunches URLs to internal twitter ones, complete with analytics.")

There is also no compelling reason for people to pay for their service or their data -- the notion that the second has any value at all appears to primarily be bubble thinking utterly disconnected to any actual experience of selling data. (A few hundred people -- can't quite tell you who, but they're out there -- like computerized voices which sound like Hannah Montana, as demonstrated by the response to this tweet. Shazam! Pay me money.)

they're primarily parasiting their users off another popular service (Twitter)

source? not sure that's totally true. (maybe for tinyurl, which twitter uses automatically, but bit.ly?)

Here's a list of their most popular links.

http://twitter.com/bitlynow

Copy/paste each link into their search page, then click "info", and you'll see where the users are clicking them.

No - I'm just going to take your word for it :) Thanks!

Edit: oh wait. In the top 10:

Email Clients, IM, AIR Apps, and Direct

Direct Traffic includes people clicking a bit.ly link from:

- Desktop email clients like Microsoft Outlook or Apple mail

- AIR applications like Tweetdeck or Twirhl

- Mobile apps like Twitterific or Blackberry Mail

- Chat apps like AIM

- SMS/MMS messages

It also includes people who typed a bit.ly link directly into their browser

seems to be 10x of twitter.com, so it's not easy to say that most of that is API traffic from Twitter clients, then.

So, then, no substantial evidence. I'm just curious whether it is a good assumption that twitter provides most of the traffic for all URL shortners, or not. Perhaps the people that run URL shortners can comment?

Yep, I guess his point is that someone who only understands code to be a barrier to entry is not really informed enough to discuss the matter properly. As you point out, the important thing is to have some barrier to entry for the competition.

The difference between Twitter and bit.ly is that Twitter users greatly benefit from the network effects of having all the people they care about on Twitter.

With bit.ly it doesn't matter at all if I use the same URL shortening service they use or not. The second something better comes along, or Twitter decides to switch to something else, or they attempt to monetize it by inserting ads or something there will be a mass exodus.

Of course there's always the possibility that the bit.ly folks pitched some grand vision we don't know about to their investors.

On a related note, I have a feeling our reliance on URL shortening services is going to bite us in the ass in the future. When one of these service goes out of business you're left with thousands or millions of essentially dangling pointers. It also throws a monkey wrench into PageRank type algorithms.

@PageRank: Doesn't Google already follow redirects? (Most of) those URL shorteners aren't doing anything else.

More to the point: Technical sophistication is one way of keeping ahead of the competition, and probably the most appealing to hackers; but it's not the only one. You can be very successful selling complete garbage; just look at Microsoft.

Edit: Yes, I'm being facetious in poking fun at Microsoft.

That Microsoft comment is both trite and undeserved. Microsoft doesn't particularly sell "garbage"; no, it undercuts application vendors for its platform, copies their products, delivers a freebie/cheap knockoff to the market to compete with them and drive them out, but works very hard to improve it as time goes.

Microsoft is evil but it's not incompetent. Sometimes it even puts its market dominance to good use; MS whipped hardware vendors into shape and gave us a truly uniform PC platform with Plug and Play.

Also, many Microsoft products are actually quality products. The few bad apples give the rest a bad name (also, some of the 3rd party apps give Windows in general a bad name).

Agreed. The UI for the desktop just gives me a headache but there are some really cool projects coming out of Redmond. There is some really interesting work in computational photography:

http://research.microsoft.com/en-us/um/cambridge/projects/au...

http://livelabs.com/photosynth/

I wouldn't give up my Mac or my Linux PC for anything, but over the years I've found that Microsoft products are vastly superior to what they were circa 2000.

BTW, you can blame Microsoft all you want, but they're the ones holding 90% of the desktop market, not our Macs and Linux PCs. Their consumer products might not be top-notch, but they did many things right - things which Apple and Sun and Linux vendors failed to do.

One of the central points in "20 years of high tech marketing disasters" is not that Microsoft are so smart, but that they had fewer major screwups than everyone else. I think it's a sensible point of view and highly recommend the book.

My summary: http://www.squeezedbooks.com/book/show/10/in-search-of-stupi...

The twitter comparison isn't really accurate - if you & the people you follow are on twitter, it's not trivial to just switch to identi.ca. Switching from tinyurl to is.gd to bit.ly to whatever comes next is a lot simpler.

.LY is for Lybia, what if the nutjob that rules the country decides to shut down their internets?

A billion links linking to nowhere...

Not long ago, Libya was on US trade-restricted lists.

If Libya were to return to such lists, would using a bit.ly URL to link to advanced encryption code be a federal offense?

Reply on news.ycombinator.com