This guy is completely mistaken. It did at least occur to him that the two most successful web startups to date, Google and Yahoo, both did what he says startups shouldn't. He replies to that by saying that you can only get away with a goofy name if people love what you build.
If he had followed that line of reasoning one step further, it might also have occurred to him that, conversely, if you make something users hate, it won't matter that you have a good name.
In other words, what matters is what you build, not what you call it. Names just don't matter much. Which means he's wrong, since the thesis of this article is that they do.
Startups choosing random names are in fact pursuing a near optimal strategy: just choose something reasonably short and catchy, and spend your time working on the product, because that's what matters.
Pogue isn't arguing that names are life and death for a startup, he's just pointing out the obvious: Given two sites that do reasonably similar things, the one with the "catchy" name is going to be remembered.
I think there is a tendency here to conflate short with catchy. In my opinion, there's nothing catchy at all about any of these: Etelos. Iyogi. Oyogi. Qoop. Ooma. Some of it may be a function of personal taste, but I think there's plenty of room for improvement.
No. If you read the article, he's complaining about names that are merely catchy-- which is exactly what Dr. Seuss words are: interesting sounding, but "meaningless" and "silly."
His claim that names matter for a startup is implicit in choosing to write a whole article about how the names startups currently choose aren't good enough.
I did read the article, and what he's really complaining about is words that are "meaningless" and "silly", and still are not "catchy" (oxford: instantly appealing and memorable). He's most concerned about the fact that the names are not memorable.
There is obviously an implied claim that startup names matter to a degree, but I think he makes it clear he doesn't think it's priority number one. Not to mention, it wasn't a whole article, it was a blog post. He writes for the Times, and he certainly didn't choose to make this the topic of one of his printed articles. He certainly isn't the first, nor will he be the last, to complain in his blog that web startup names these days kind of suck. I'm not naive enough to think David Pogue's blog is anything like mine or most, but it's useful to gain some perspective.
I don't think he understands the extent of the availability problem. We wanted to avoid the Web 2.0-ish name, since our audience is a little more pedestrian, and ended up having to write a script to combine keywords and check for availability.
And there were certain keywords for which there was just no good option left. For instance, sports. Our keyword list returned thousands of total availables, only two of which contained the word sports. One of which was sportshit.com.
Naming probably mattered more in the era of phone books and neon signs: if you didn't recognize the name as relating to the service you wanted, it counted against the name.
These limitations aren't present with a URL; a short meaningless name is probably better than a long meaningful one.
An interesting recent phenomena is companies replacing their trusted names built up over many decades with a short, catchy, meaningless one: the rebranding of Washington Mutual to WaMu comes to mind.
I agree with Matt. If you are going to use a word from another language, don't do it because it means something relevant to your business. The only people that will know it means something are people you meet and share the story of your name with. Vast majority of your customers will have no idea of your name's meaning.
ie. last year over summer vacation I would see kijiji ads behind rickshaws all over Bombay. Only now I know it means something. Before this, I always thought of it as a lame name. Knowing it means something makes it slightly less lame--but not a whole lot.
It's much better to go with a memorable name versus a cryptic name from another language.
Not really. Words from such languages are commonly found in the West. Microsoft named their Silverlight search front end Tafiti which is also derived from Kiswahili (tafuta - to search). Kiswahili is a widely spoken language in East Africa. (http://en.wikipedia.org/wiki/Swahili_language). There are many other occurrences of the language in popular American culture! It makes sense to borrow words from other languages when there is "just no good option left" as you yourself say.
Tafiti is gibberish too. Any word I haven't heard and can't squarely place the language of is gibberish to me. And given that like 8 people who own a computer have ever even heard of Kiswahili, Kijiji is gibberish.
If he had followed that line of reasoning one step further, it might also have occurred to him that, conversely, if you make something users hate, it won't matter that you have a good name.
In other words, what matters is what you build, not what you call it. Names just don't matter much. Which means he's wrong, since the thesis of this article is that they do.
Startups choosing random names are in fact pursuing a near optimal strategy: just choose something reasonably short and catchy, and spend your time working on the product, because that's what matters.