Sixty eight weeks?!
I think there’s a bigger point to be made.
It’s easy to assess things that we are familiar with, but it is impossible to assess things that we cannot appreciate. Therefore it’s best to learn a little about a lot and assume you’re ignorant until you can prove yourself otherwise.
the “stackoverflow-weekend challenge” is definately a prime example of the 80/20 rule… or so i would imagine
Jeff – I notices the link to serverfault is broken. Looks like you made a typo (serverfault.co,)
I only noticed cus it said I hadnt been there…which is definitely not true!
I never cease to be amazed at the number of working software developers who devalue their own work by insisting that all software should be free. That comment about SO goes one step further and presupposes that programming is monkey work and therefore worthless.
When people start taking cheap shots like that, you know you are on the way up. Take it as a compliment.
I wish it were only developers that thought this way. The number of prospective clients that arrive at my web development agency with the same mindset is slightly depressing. There really is no understanding that a good user experience takes time to put together, and just how much time and effort companies (like StackOverflow, Google, Facebook, Yahoo etc) have put into the user experience.
Other developers having this mindset I can understand - it’s a common problem as you point out. Non-technical users… tell me where they get the idea from??
Getting the site functional is one thing but tuning it is quite another. Adjusting css classes by a pixel until things are perfectly aligned. Taking into account differences in browsers. Trashing a page’s layout because what looked good yesterday suddenly today looks like your little brother got a hold of it. Those are the things that really take a long time. I think a good developer is never really done with a project.
@Brian, “Get over yourself. SO is not a complicated website to recreate. I do think development of it would take more than a weekend however; likely, 4.”
Its easy to “recreate” something after something has already been created. Coding the website is the easy part. Coming up with the ideas is the hard part.
Looking over the StackExchange page, there’s one thing that sets my alarm bells off. For all but the most heavy duty of plans, you do the hosting. But there’s nothing promising that my data is mine.
If you don’t make it possible to get your data off of StackExchange then you’re leaving a huge window of opportunity for an open source implementation. While few companies would give up a polished, debugged product just to save the monthly fees many would give up a polished, debugged product to know that they own their data.
There’s nothing scarier than thinking you’ve spent several years building up a knowledge base that you now have to pay perpetually to access.
Spot on!
In some sense, this is inevitable. To be a good programmer, it helps to be more comfortable with computers than with people. Otherwise, you’re unlikely to spend the necessary thousands of hours hacking alone.
The flip side is that many programmers know the inner workings of people about as well as J. Random User knows about microprocessors. That was certainly true for me well into my 20s.
It’s an easy leap from not knowing much to thinking there’s not much to know.
When smart programmers without social smartness look at a site getting profitable by day, only thing they can thing is “I could have done this”!
So, let’s move on.
So programmers are easily replaceable by monkeys, but the only real work is done by the CSS/JS/HCI guys?
Good post. Agree with pretty much all of it.
Apart from the points raised in this post, I also think it’s indicative of the open source community that they would sooner be cloning other peoples successful ideas then coming up with something original.
One point though. How did you manage to go from “Open source software only comes in one edition: awesome.”, to “open-source software remains such a horrible pain in the ass to use” in five days?
I think the illusion of triviality is essential for people to get started on the project and create their first hack up - in the spirit of “don’t worry be crappy”. If we do really understand the full extent of the work involved, we may never start on a path that can turn out to be a fascinating journey. … as the Twitter folks probably discovered 
As it happens, I’m not interested in hosting my own SO clone, I was merely being hypothetical. But in response to your question - if I WAS hosting an SO clone, I would hope it was attached to a genuine business model that would provide the required income, thus I wouldn’t need the ‘frippery’ of ad-support.
Running a business on ad revenue is a mug’s game - Joel clearly understands this, hence the StackExchange model. Every dotcom I’ve worked for has had a real business model and buys ads rather than selling them.
Writing code is always easy, but StackOverflow (and twitter and facebook) are successful because of the community that uses them. You can’t build a vibrant active community in a weekend. Jeff and Joel have been working for years creating communities that they can leverage to make StackOverflow successful. StackOverflow could be horrible, slow, and buggy code and you’d still be successful because you have a great following of people who love to talk about how smart they are.
We all could do SO for less than a weekend. But design = 1 month, making sure buttons click the right way with pleasant animation and all other UI tweaks you cant even imagine = 11 more months. And when I try to sell it = no community, untested scalability vs Jeff’s already popular product = total fail. the only way I can beat it is by distributing it like fogbugz - here is the code, send back the tweaks and bugfixes, pay some one time fee (My company is in need of SO clone at the moment,but providing only hosting solution is a big showstopper).
the missing login / logout kepps me away from using so.
think you need more polish.
The missing login keeps me away from so.
Bette rpolish it up…
and for stupidity:
Coding Horror
programming and human factors - Jeff Atwood
Comment Submission Error
Your comment submission failed for the following reasons:
Publish failed: Renaming tempfile ‘C:\codinghorror\blog\archives\001284.html.new’ failed: Renaming ‘C:\codinghorror\blog\archives\001284.html.new’ to ‘C:\codinghorror\blog\archives\001284.html’ failed: Permission denied
Return to the original entry.
“There’s nothing particularly magical about the production of source code. In fact, writing code is a tiny proportion of what makes most businesses successful.”
If this is, in fact, the case, then it can’t hurt to open source your software’s source code so that people can make good use of it, recognizing that (a) the real work in competing with you on a business level has only a tiny bit in common with your actual code and (b) the follow-on advertising and awareness you might get from others using your source might even feed back into making your own product more successful.