Programmers think:
success = lots_of_awesome_code;
not:
lots_of_awesome_code = success;
Joshua, yes, but is it…
success++
or
++success
?
best quote: “Not to write the best Stack Overflow code possible, but to create the best Stack Overflow experience possible”
Some of the daily deal sites I visit just toss together some opensource forum software in an attempt to build ‘community’, but woot wins hands down with even their modest ‘forum’ engine.
Maybe what the ‘trivial’ comment implied was that they could get something ‘started’ in a weekend. Is it ever reallly finished?
You left out Benjamin’s final grandeurposturing conclusion of “That is why an open-source clone of StackOverflow will fail.”
Twitter comment was hilarious - perfect touch to the end of a post like that.
Amen!
More to the point, I was once asked how long it would take to create a line of business handheld app. None of us had any experience in making these types of apps and I could easily recognize that this app needed major architecture work upfront to meet the long term goals of the project. My response was that it would take 1200 hours. The higher ups scoffed at this and I was sent back to my cubicle. A young programmer with significantly less experience in handheld apps proclaimed that he could easily do it in 200 hours. And so he was given the project.
1200 hours later he was finally done. But only after the customer had been put through hell. I had since left the company but was having lunch one day with my former coworkers. One of them had mentioned to my boss “Gee, I guess Matt was right afterall!” to which he had reluctantly agreed.
The “I can do that in a week” mentality is all too prevalent in our world.
Actually, the “polish” isnt contrived, nor is it something provided in Drupal. If you think so, I think you are missing the point of Jeff’s blog entirelly. What makes StackOverflow isnt just the software, its also the effort put into the experience, and the effort put into the marketing. In short, SO is a business, not a software product. Even if you had the software, a literal exact copy of what SO has, you would be able to reproduce SO without a business plan and a good marketing strategy.
Hacker News is full of young bucks who are straight out of school and full of piss and vinegar. I know, I used to be one of them.
When you’re young you don’t think that time and money are fungible because you have more time than you know what to do with and very little cash on hand. Therefore the idea of paying for stack exchange seems silly when you could just sacrifice forty hours(or more) of your life and get it for free.
As you grow older, and especially once you have kids, time and money become much more fungible. I’d rather pay stackexchange a grand than spend a weekend away from my kids writing code that’s readily available and fully functioning.
No, I don’t take this claim seriously. Not enough to write a response.
Oh, and what this post was about then?
Just shut up and do it if you think you can. I’ll come use it if it’s better.
Good luck.
Well, I find no convincing reason for anybody going for an Open Sourced clone of SO. Considering that the question and answers of SO are available under CC license, I do not see the necessity for an OS community to go and write a clone of SO, as what they community is going to care is whether the knowledge part of SO which is in the form of questions asked and answers given is Open or not.
Secondly, writing a SO clone is not going to anyway hit the business because for a SO clone to raise itself to the level of SO it has to travel miles. Also, a mere software clone really doesn’t help much, it also needs the reputation which SO has earned through all these time.
I guess code is trivial but only as trivial as anything else is. I know when I have to mow the lawn I have to get the mower out, check the oil, pick up everything in the yard while checking for stones or soemthing that may fly out and kill a kid or ding a car. Start the stupid thing and then decide how I am going to go over the lawn this time. Then I have to wack the weeds and think about putting something on the lawn to make it look good.
But mowing is trivial as well. The difference to anyone is how much you enjoy it or how much you dislike it. But you are right on with the fact that anything done well is done in a very small place. Success is made on tiny edges. Everyone has the same resources to get things done for the most part. Making small improvements, better distinctions is what makes a product good.
This is why people don’t pay for ideas. Ideas aren’t worth anything until they are actually implemented and working. So saying anyone could code up Stackoverflow in a weekend isn’t really saying anything until they have done it.
On the other hand, this comment posting system could use some work. Apparently file copying errors still post but don’t show up right away, leading to my double post. :^P
@Michael: Wow, after a brief look at that site I can say… it makes a GREAT example of what Jeff is talking about with regards to polish. Sure, you can ask questions and get an answer there… the rest of the SO experience is missing. User ratings, badges, usability, pleasing design.
It’s an example of what you can churn out in a week, and what took the SO team another year (or however long) to turn into what SO is now.
@Russ: You wouldn’t care about the ads? They aren’t an important part of the site? Wow, I hope you expect to lose LOTS of money when you run your SO clone and pay all the bandwidth charges without any actual income!
“Although, to be fair, I really could write Twitter in a week. It’s so ridiculously simple! Come on!”
If you just made a fail-whale webpage, it would be right about 20 minutes a day!
It takes a lot to bring code down to the trivial level. Most developers tend to complicate things. It is just difficult to tell people they do not have to write tons of code to make things work.
I say Kudos to SO.
The correct estimate is, of course, six to eight weeks.
The word “doom” is spelled “how hard can it be?”
Agree with Pablo or Joshua. It should be
lots_of_awesome_code == success;
or at least
success = lots_of_awesome_code;
instead of
lots_of_awesome_code = success;
Sorry for nitpicking…