Consumers can get over a weird name (Wii?!) but what hurts is when they can never remember it.<p>It's easier to remember names that are concrete (even though monkeys have nothing to do with surveys) or have a memorable connection to the product's usage, like Reddit.<p>When someone hears the name, understands the product's use or sees the site, try to give them a thread that they can pull on later to remember the name. That's all.<p>I feel bad about calling out Ninite as a counterexample but several HNers say they can never remember their name. If your name isn't relevant, at least let it be memorable.
Though I liked the blog post, I think that the conclusion is a bit facile. The analogy that comes to mind is "going to the store to buy a new PC and coming home with a new keyboard instead for your existing one".<p>Sometimes you don't appreciate what you have (with minor modifications) until you take the journey.
It sounds like a good amount of time was wasted because the domains were already taken, which was only discovered after discussing and voting on them. It might have been better to forfeit all names without domain in first place. It's still possible to find good, free names. <a href="http://impossibility.org" rel="nofollow">http://impossibility.org</a> is a pretty good domain name generator, for instance.
This reminds me of something that happened here in Singapore not too long ago. The government had big plans of reviving the downtown area with lots of new developments, including the new casinos. As part of this makeover, they thought the then name of the area - Marina Bay - needed to be retired too.<p>So they hired a consulting firm and paid them about half a million dollars (IIRC) to come up with a new name.<p>In the end, the name they picked was.. Marina Bay. Yes, half a million to just continue with the old name! The spin was that the name was good and already part of the national culture/image. Which, to the layman, was obvious from the start :-)
Just my opinion, and you're right that it pretty much doesn't matter, but I don't like "Trello" because we already have "trulia" and "twilio". They just are too similar to me.
That's all good if you are an notable evangelist like Spolsky, but i think it's best to use something at least vaguely descriptive. Sure, people can remember the 7 most popular sites on the net, but for the rest it's best to choose something that can come up easily when you struggle to remember the freaking name of that service. For example, Picplz=obvious, groupon=sort of obvious, Quora=might come up when thinking "questions". I prefer to pick the site subject and add a funny/fuzzy or non-english suffix for my projects, e.g. projectilo.com, noteplz.com, mapifier.com etc.
Naming can be an agonizing process, especially the need for domain name and SEO consideration.<p>OT. Are number in product names or domain names bad in general? E.g. foobar101.
Another interesting part of the project name / url debate:<p>Is having [product].com that important?<p>In their book and blog 37Signals has generally played down the importance of urls. Pointing to the use of basecamphq.com for their project management offering.<p>I must say I am on the fence. Clearly appending something to your product's name opens more options for un-squatted domains but at what cost to the end user?
Is it trello like hello, or trello like llello (yeyo)?<p>I dislike names which 1) can't be said and transcribed unambiguously over a 4kbps bad cellphone call by a non-native speaker to a non-native speaker and 2) don't have an unambiguous pronunciation...people might not mention your product if they're afraid of sounding stupid mispronouncing "ver-say-che", etc.
One aspect I'm interested in is the SEO angle. For instance, if I make cow food and call myself Cow Food Online, register a domain of cowfoodonline.com, would I stand to benefit from having lots of links with "Cow Food Online" pointing at my domain, since that'd be the name of my service?
The post also mentioned "Idealist". Would have been a much greater name, lots of clever wordplays and explanations what it really is: Idea List Ideal Deal. I wonder why they decided against it.
One thing we learned is picking a good name is something you should do from the start. We spent considerable time and energy replacing our earlier terrible name (TransFS) with one we like (FeeFighters). The process was pretty similar to Fog Creek's... we went through a huge exercise, met with branding firms, etc. At the end of the day it was an 12th hour "we need something" decision and now we love it. Who knows, we might have loved bidsquid and costhammer just as much...<p>More on that in this Mixergy interview: <a href="http://mixergy.com/sean-harper-feefighters-intervie/" rel="nofollow">http://mixergy.com/sean-harper-feefighters-intervie/</a>
Yes, naming is a huge time-suck. Not only do you have to worry about the domain and the github project name, but you have to worry about trademarks as well. Even if you hate trademarks and have no intention of registering yourself, some other Johnny-come-lately could register a name you've worked hard to give meaning to and then deny you its use. Prior use is an even less effective defense with trademarks than with patents, and international trademark issues are too nightmarish to talk about.<p>Take it from someone who has actually had to change a project name after two years. You want a name that's pronouncable, barely memorable, but nothing to do with any real words that others might contend for. There's a reason for all those crappy nonsense-word names. They put an end to the pain so that developers can go back to developing actual functionality.