What We Can Learn From Mess (known as Craigslist) | Hacker News Reader

What We Can Learn From Mess (known as Craigslist)

avc.com

41 pointscwan27 comments

Article headline title should be:

What We Can Learn From Elegance

Craigslist also wins by having the right categories, right font size, right colors. It is fast and the urls are predictable. Also, breadcrumb nav and search is good.

It was precisely what I was thinking when designing http://twitya.com - a twitter browser.

When you visit a site every day, it should be as fancy as your arm, not unlike this site.

Content beats presentation? Staying close to the customer rules? No meetings a big win? Worthy of contemplation.

These are principles many of us have been telling bosses, clients, readership -- anyone who would listen to us -- for quite a while. They're principles that get ignored every time they're brought up, though. It's awesome that Craigslist is demonstrating that they work, and terribly sad that so few people seem to be taking notice.

Why sad? That's more opportunity for those of us who do take notice.

Do you not find the notion that humanity is full of ignoramuses "sad"?

Gaining advantage because the rest of the market is populated by idiots doesn't make it any less sad that everyone out there is an idiot.

Instead of assuming everyone is idiots, try assuming you need to gather more convincing evidence to support your ideas. It might not be more correct but it will probably be more productive.

These two concepts -- that many people are ignoramuses, and that more convincing evidence might help -- are not exclusive of each other.

No, but the assumption of the ignorance of others as the reason you can't reach them is passive whilst the drive to provide better evidence to convince people is active. Sitting back and saying I've got what you need but you can't understand is not going to move you forward. Pushing on to a position of greater clarity and persuasion will usually help in convincing not only the as-yet-unconvinced but also win more mindshare with the convinced.

[Was that generic enough?!]

I don't see where you get the impression that I'm advocating for (solely) taking your "passive" approach, either. Against what, exactly, are you arguing? Is there something I actually said to which you object? Are you assuming for some reason that the fact I said something is proof that I advocate everyone doing nothing but sitting around saying the same thing? I even agreed that providing more convincing evidence could be helpful, indicating that trying to come up with that evidence is a good idea! What's up with your attempts to cast everything I say as opposition to that?

I was remarking, rather than arguing. More of an expansion of my interpretation of the GP's post. Note that I start "no" as agreement that the two things are not exclusive.

[I think if we're not on the same page then we're at least in the same Dewey classification.]

Okay, fair enough. Thanks for clarifying.

Extrapolating from "it worked for Craigslist" to "it will work for us" is highly naive. If someone showed up to a meeting and said, "let's do like Craigslist" and totally ignored all the ways that our business is very different from craigslist I'd be sceptical too.

Who said anything about "extrapolating from 'it worked for Craigslist'"? I didn't say "Some of us have been trying to get our bosses to copy Craigslist for years!" What I said was that some of use have been aware of such principles for years, and telling others about them. One hopes that sharing such principles would involve applying them properly, rather than naively trying to copy someone else's formula.

Do you just dislike Craigslist, and thus hate anything related to it -- including its principles of success and anyone who happens to be aware of some of those principles?

I, for one, have never told anyone "You should base your business model on Craigslist's model!" but I have said "You should probably simplify your business model. I have some suggestions."

My wife and I run a business together - we practically live in each others pockets. The business is small in every sense and yet we still need to have [occassional] meetings.

I don't believe they don't have any meetings, maybe they moved their meetings online, but they'd still be meetings.

Obligatory reference: http://www.aperfectmess.com/ a book about the cost of neatness. And this debate covers a lot of areas. Anyone that trusts science must recognize the power of emergent patterns, and how they rule the world. This, the internets, is one of the most interesting examples of it because it was just born some years ago, world geography and big cities are as interesting but much more afflicting examples. Will there be a science of disorder? In the case of enterprise architecture and Craiglist, what are the principles? Are they embodied in Craig Newmark? Or there will never be a science of disorder, and rest of us will be left with the social sciences which try to explain how we value neatness and hoping more enlinghtned people like Craig Newmark will rule the world?

There is a breach in the Master & Commander pattern of today enterprises waiting to be opened.

Another wildly successful example of "What We Can Learn From A Mess" can be found at Plenty of Fish, where they are doing the same thing in the online dating world.

http://www.inc.com/magazine/20090101/and-the-money-comes-rol...

"Today, according to the research firm Hitwise, his creation is the largest dating website in the U.S. and quite possibly the world. Its traffic is four times that of the dating pioneer Match, which has annual revenue of $350 million and a staff that numbers in the hundreds. Until 2007, Frind had a staff of exactly zero. Today, he employs just three customer service workers, who check for spam and delete nude images from the Plenty of Fish website while Frind handles everything else."

"a staff of exactly zero" Or more correctly "Plenty of Fish had a staff of one"?

I find all of this very intriguing. Assuming that both Craigslist and PlentyOfFish are successful for the same underlying reasons, it's worth pointing out similarities between the two. Note that while I've been using Craigslist for years, I have very little real experience with PlentyOfFish, so feel free to correct any invalid assumptions I make:

1. Plain user interfaces. The choices of fonts and colors seems very personal, non-corporate. There's nothing sleek, modern, trendy, creative, or flashy about either of these sites. That's not to say they're particularly simple or efficient.

2. Unchanging user interfaces. When you go to Craigslist, at least, you can be sure to find the exact same site you were using one, two, three years ago. No surprises, no re-learning, and no feeling left-behind.

3. Initial focus on small community, growth through word of mouth. Craigslist had San Francisco, and PoF had Vancouver I believe? There's something to be said about making users feel like they are party of a small, exclusive community. (Look at the beginnings of Facebook.)

4. Content created almost entirely by users. Craigslist does little more than categorize users' posts, and PoF is, well, a dating site.

5. Open: you don't have to know someone or get them to accept a friend request in order to interact with them. Once again, these sites make connecting with others easy and direct. (Although the signup process at PoF is understandably more involved.) When I go to Craigslist Boston I really do feel like the site represents the entire community.

6. Pretty much free.

I'm sure I could add more, but the recurrent theme seems to be getting out of the way and keeping the focus on the people. Neither of these sites rely on revolutionary tools or features or algorithms, and neither seems particularly interested in impressing the tech community.

Another one to add to the list is CD Baby (www.cdbaby.com), a site owned and created almost entirely by Derek Sivers. He sold it for $22 million last year.

Looking at the site now, it's a bit sleeker and more up-to-date than both Craigslist and PlentyOfFish, but not too long ago it looked pretty much like a link farm: http://www.crunchbase.com/assets/images/original/0004/6518/4...

It seems like people are making the mistake of generalizing from a sample of one. Craigslist does things in a certain way, and craigslist is a success. The latter doesn't necessarily follow from the former.

Sure, it's a lean organization, and sure, that's good. They've cut out a lot of stuff that they can truly get along fine without. But if craigslist didn't exist and you wanted to launch it today, you would never launch the design it's famous for. And yet you'd stand a good shot at real success.

So craigslist has been successful in spite of its design. In the face of the things people like about the site, the design ultimately didn't really matter. I don't know why people can't just leave it there and instead feel the need to infer causation.

To infer the opposite truth without investigation or evidence to back it would be equally wrong.

I would put Craigslist's clean/basic design in the list of reasons why it succeeded while others failed. It may be a small % but they certainly didn't succeed despite it.

It is a clean design, and that's good, but there's a really long list of things that are wrong with it. Things they could fix while retaining a clean, stripped-down experience. In light of that, I'm confused about why people ascribe so much of the site's success to its design. A better design doesn't have to (and shouldn't) mean a heavier one.

It doesn't seem to have hampered craigslist, though, and that's why my best guess puts it in the "nonfactor" category.

None of what you say Craigslist did wrong obviates the importance of recognizing what Craigslist did right. It seems that what it did right has proven more important than what it did wrong in this case, in fact, further attesting to the value of what it did right.

If you want to do a similar job better than Craigslist did, you should probably do what Craigslist did right, and look for ways to improve it without undercutting those positives. In short, the fact Craigslist does some things wrong doesn't mean you can't learn from its example.

> So craigslist has been successful in spite of its design.

How about some supporting evidence?

Better yet, how about some evidence for your belief that certain changes would make it more successful?

Of course, such evidence should include some indication of how we're measuring "success".

Craigslist succeeded because there is an ecosystem of people that makes it useful. buyers/sellers, job hunters/head hunters and so on. The same reason eBay is successful.

But Craigslist beat others in establishing theirselves as the hub of that ecosystems; they didn't just buy the domain name and have all those groups of people there the next day.

I agree. first-mover advantage plays a big role as well.

Reply on news.ycombinator.com