Wow, the Cuil buzz died quickly. In code...

20 points by shafqat ↗ HN
switch ($day)

{

   case 1: 
      echo "All the rage!";
      break;

   case 2: 
      echo "Worst product e-v-e-r";
      break;

   case 3: 
      echo " ";
      break;

}

37 comments

[ 3.5 ms ] story [ 74.1 ms ] thread
You really should encapsulate string generation in a function.
no default case?
not having a default case actually makes a message: silence. as in I already forgot about cuil because the media is silent about it.
Can you mark the strings as translatable and send me the .po file? thanks.
I prefer:

  cuilBuzz = {
      '1': 'All the rage!',
      '2': 'Worst product e-v-e-r",
      '3': None
  }

  day = 3
  cuilBuzz[day]
It is weird and magical to me that the lisp guy got modded up but the JavaScript guy got modded down.
You'd think it's JavaScript, but actually, it's Python.
(comment deleted)
Maybe he got modded down for posting code that won't run.
I wrote it in Python, actually. And it has some very obvious bugs. Here's the revised code:

  cuilBuzz = {
      '1': 'All the rage!',
      '2': 'Worst product e-v-e-r',
      '3': None
  }

  day = '3'
  print cuilBuzz[day]
  # Although the print isn't necessary...
Now what would be interesting would be an accurate version using the datetime module...
This is my preferred method for Python (faux-)switches:

  cuil_buzz = {
      1: 'All the rage!',
      2: 'Worst product e-v-e-r.',
      3: ' ... '
  }.get(day, None)
As sad as this is, it's up to their marketing department now :-P
Very good, but you left out step zero, in which a PR firm is given a large check.

It's hard to argue that they didn't earn the money. Whether that money was well spent is another question.

I know I shouldn't but I can't help it:

  ;; format magic
  (format t "~[~;All the rage!~;Worst product e-v-e-r~; ~]" day)
  ;; a little less magic
  (format t "~A" (case day
      (1 "All the rage!")
      (2 "Worst product e-v-e-r")
      (3 " ")))
Yay for Common Lisp!
I think this may highlight the risk of a "Bang" release. Sure, you get a ton of attention all at the same time, but if your product isn't good enough, that's going to be hard to fix (The impression in peoples minds).

I wonder why they didn't just show it to a few hundred people first... maybe they did, I don't know, it just seems like a bit of a bad way to release a product to me.

build product in a vacuum -> release it to the world!!! -> Oh... umm... they don't like it

I assure you that the buzz has not died down for people considering serious investment in Cuil. And ultimately, that's why people do buzz launches of demo products like Cuil (and Powerset before it) did.

The mandate is to prove:

1) That you can make an infrastructure capable of generating indices and hosting a search site that will get major traffic.

2) That your product is not vaporware.

3) That your product has room to grow in terms of technical capability, to justify why a company is seeing investment.

Okay, I spent the last 15 minutes writing the code above in assembler, and then looked at myself and thought - what the hell are you doing?
Getting your geek on! Finish it!!
Up, up, down, down, left, right, half-circle punch!
Powerful you have become, the dark side I sense in you...
Not following through has a certain 'authenticity' and honesty.

    day = 15

    case day
      when 1 then puts "All the rage!"
      when 2 then puts "Worst product e-v-e-r"
      when 3..(1.0/0) then puts "...uhhh...nil, but Ruby still loves cuil, even if others dont!"
    end
That's the nature of buzz that gets you in front of 1/200th of internet users. (Based on Alexa stats, at least.) Also according to Alexa, they are only down about 4x from their peak at launch. Cuil claims they got 50 million searches on launch day, so if this logic is even remotely close to correct they are still seeing a good amount of use.

Of course, each of these stats sources are somewhat unreliable, and combining them raises its own issues. Caveat emptor. ;-)

I bet that in a month, Cuil is still a top 1000 site according to the usual measurement services. Maybe the buzz has died and I find many faults with their launch strategy (and results) -- but they still built and shipped a fast Internet-scale search engine. That alone should give them a lot of slack.
switch ($day)

{

   case 1: 
      echo "All the rage!";
      break;

   case 2: 
      echo "Worst product e-v-e-r";
      break;

   case 3: 
      echo " ";
      break;
   
   default:
     die('not cuil ... not cuil');
}
Wow, we are all geeks. Love it. Pleasantly surprised to see this submission get so much love!
wow I guess a lot of people never here never made a mistake before

most of you guys probably already have a very successful company by now too

Not everyone here thinks it's a mistake. There are going to be better solutions to narrowing the signal/noise ratio in the search space than Pagerank.
constructive criticism is one thing, making a big joke out of it (like most of the posts I see here) is another

I was trying to be sarcasstic with my 1st post, I guess it didn't work

i'm not sure people are making a big joke about it. i think we all learned something from the cuil launch. as i have said before, only time will tell and we should give them the benefit of the doubt.

i made the first post, and it certainly wasn't to make fun of them. apologies if that is what it sounded like.

http://paste.lisp.org/display/65257

Brainfuck generated from:

    ,48i-:1i=["All the rage!"s[.]_10i.!]_:2i=["Worst product e-v-e-r"s[.]_10i.!]_:3i=[32i.10i.!]
Funny, I wrote the language yesterday to familiarize myself with Parsec, didn't know I'd find a use for it so fast. The day has to be entered manually.
what happens on day > 3?
default: die( "Goodbye Cuil World!" );