Wretched Word of the Week: Killer
The word “killer” gets used in two wretched ways. The first is Killer Application, and the second is product-killer. They’re each wretched in their own special way. It’s not only cliché to use each term, but in using it, you are nearly guaranteed to be wrong.
The original killer application was Lotus 1, 2, 3. It was the killer application because it was the application that made early PCs desirable by large numbers of businesspeople with budgetary authority. They bought 1, 2, 3 and the PC was a means to that end.
Arguably, there hasn’t been another killer application since Lotus. All the ones I think of are diminished in scope. Killer applications are things that appear once in a very long while, and appear when the underlying thing they promote is immature. I can make the case for some uses of the term. It’s been said (and I have said) that email is the killer application of the Internet. Certainly, many people got on the Internet (or stayed on it) because of email, even more than web browsing, even through web browsing gets all the press. One could argue that TiVo was the killer app for satellite TV, or for something. There’s an old MIT saying that if you change something quantitatively by an order of magnitude, you change it qualitatively as well. I think that TiVo (and its brethren like Replay) is the VCR so improved that it’s a new thing.
Nonetheless, killer applications are once-in-a-generation thing. If you see an article that asks, “Will Foo be the Next Killer App?” the answer is almost certainly no. Killer apps are like porn and art. You know one when you see it. If you have to ask, it’s not going to happen — or perhaps it’s better for me to say that betting against is the smart bet.
The second wretched use of “killer” is the product-killer. I’ve been sitting on this post for a couple weeks waiting for someone to write some article about an iPhone-Killer, and Michael Calore of Wired News wins the prize for his, “The $300 Linux-Powered ‘iPhone Killer’ Arrives” [link to http://blog.wired.com/monkeybites/2007/07/the-linux-power.html no longer works].
Mind you, I’m as sick of iPhone hype and anti-hype as the next person, and I think that OpenMoko is pretty cool. However, an OpenMoko phone is no more of a iPhone-killer than a different Lotus, the Caterham 7 [link to http://www.caterham.co.uk/assets/html/showroom/build_manual.html no longer works], is a Ferrari-killer. It’s not an insult to the Caterham to say it’s not a Ferrari-killer. It’s no insult that OpenMoko is no iPhone-killer.
Like the Ferrari, the iPhone can’t live up to its reputation and generates a counter-reputation.
But just as importantly, the sort of people who want a Caterham are in general not the same sort of people who lust after Ferraris. The fun workhorse for people who take as much joy in the tinkering as the actual use has a different aura, no less powerful than glitz, but different. Let’s face it, the sort of people who would buy an OpenMoko phone are in general not the sort of people who want an iPhone.
We would expect of sports car columnists that they could tell tell the difference between a Caterham and a Ferrari. I don’t think it’s unreasonable to expect a technology columnist to be able to tell the difference between a cool Linux kit-phone and an iPhone.
Despite the headline, Calore can tell the difference. He gives you an honest assessment of the gotchas:
Keep in mind that this unit (the GTA01) was pushed out early so developers could begin writing device drivers, custom GUIs and some cool apps for the phone. The next revision (GTA02), which will be available starting at $450 in October, will be ready for the mass market. It will have wi-fi, 3-D motion sensors and added graphics accelerators. So this phone isn’t exactly an iPhone killer — the next one will be a contender. AptUsTech has a nice comparison of the NEO 1973 and the iPhone.
If you go look at the comparison, there are a number of techno-lust disappointments for an iPhone-killer. The OpenMoko phone has a GPS and twice the pixels as the iPhone, but no camera, no accelerometer, no WiFi, a CPU running at less than half the speed of the iPhone, and a piddling 64MB of flash. Yes, it takes a microSD card, but give me a break! The suffix “-killer” has been used, and I expect to say, “oh, never mind” not merely look at a “contender.” Yes, many things are coming in the next version of the OpenMoko, but this is comparing apples to orange futures.
(Also note that it isn’t exactly $300, either. The OpenMoko people have executed a sweet marketing coup in making a $300 base model that no one in their right mind would want so they can stress the low price.)
Now, in Calore’s defense, he said in his article that it’s not a killer. In short, the headline is — well — a lie. It isn’t exactly an iPhone-killer, and he’s one of the few people to point out that it isn’t exactly $300.
In more of his defense, I’m quite sure that the real guilty party here is the editor who took his article that says, “isn’t exactly an iPhone-killer” and pointed out that the phone you really want is going to be delivered in a few months for $450, and then created an attention-getting but false headline on it. Let’s face it, when you try to be nice to some cool little guys while keeping your journalistic integrity and that smacks up against ad revenue, guess which one wins?
And that is why, Gentle Reader, we should shun terms like -killer. It’s cliché, and so cliché that its negation gets twisted into the positive. I don’t know why it is that editors have a predilection for this, but it’s happened to me, too. Write something saying that the sky is blue, and you’ll see the headline saying the sky is green. All you can do is shrug and resolve to write better.
Photo of a Ford GT40 by dacorsa.net [link to http://flickr.com/photos/ijsselstein/ no longer works], and selected because I found it searching for “Ferrari Killer.”
Killer blog post, dude!
🙂
I laugh because I thought the same freaking thing when I saw the /. link come across my RSS reader, got a little worked up about it, but neglected to blog about it. Thank you for letting me have my therapy by proxy.
Is there anything more “killer” abused than Open Source software?
Let me offer Alex’s Things Open Source Software has Actually Killed:
1.) Traditional UNIX circa 1995. See Solaris, modern use vs. Solaris use 1997.
2.) SPARC. Remember when everyone’s web server was a SPARC 20? Maybe this is because of Intel more than Open Source – but one and two are related to be sure.
3.) Mac OS 9. Thank you BSD.
4.) Many non-IOS router operating systems. Most small network equipment pieces run an embedded Linux I’m guessing…
5.) The careers of UNIX scripters. If I had a dollar for every cool old UNIX guru from 1996 I knew wasn’t keeping up on streams and sockets and such because of Linux, well, I’d have at least enough for you and I to have a nice lunch.
6.) Some of the market for driver development.
What else am I missing?
Visicalc, no?
Wooo dude… You have things a little wrong here. The original killer application was VisiCalc. It showed the world that a personal computer could be used seriously for business. And Apple sales went through the roof as the result.
The IBM PC’s popularity for one reason and one reason only – it came from IBM. Which in those days was enough to prove that the PC was finally ready for business. Lotus 1-2-3 just went for the ride.
What, it wasn’t Miner 2049er?
Nuts.
The other thing about killer apps is you can’t predict them. A killer app is only obviously a killer app in hindsight, and as you say, they often only appear on immature or niche platforms.
A good example is the Video Toaster. An awful lot of TV production houses that never would have thought of buying an Amiga otherwise got one so they could stick a Video Toaster in it.
Killer apps are rare now for two reasons:
– It’s always been the rare application that has that kind of “gotta have it” pull. As you say, sort of a once-in-a-generation thing.
– Today, popular apps are rapidly ported to other systems, at which point no one has to go buy a new system just to run that app.
The exception might be game consoles. Game consoles are all about killer apps — every console company hopes someone will write one singularly awesome game that only runs on their system, and that people will buy it just to play. A console is regarded as a bit of a failure if a killer game for it doesn’t appear.
I can’t disagree too much with people who bring up VisiCalc. But I mention 1,2,3 because I was a Visicalc guy who saw 1,2,3 be the real killer app that blew VisiCalc (and the Apple ][) off the map.
Gentlepersons can disagree, so I’ll just put this link in from Etech’s “First Movers That Flopped” listing VisiCalc as a first mover that should have and didn’t [http://etech.eweek.com/slideshow/index.php?directory=first_movers¤tPic=3 no longer works].
Point taken. Reasonable people can disagree. 1,2,3 was pretty damn cool (as a Quattro guy, it is hard for me to say that).
The term ‘killer application’ was applied to Visicalc before Lotus existed. Without Visicalc there would never have been an IBM PC, middle managers would never have bought a word processor for their own use. Typing was for secretaries.
Lotus only got in the game because Kapor had written a graphing add-on to Visicalc on the PET/Apple/Tandy platforms of the day. Visicorp did not see the value of producing a version for the IBM PC, Kapor could not sell his add-on. So he wrote a combined spreadsheet/graphics program for the IBM PC and ate Visicorp’s lunch, dinner and breakfast.
And ten years or so later, Lotus gave the spreadsheet business away to Microsoft when they were too stupid to produce a version for either Windows or OS/2. The rivers of gold comming into Lotus could have easily paid for both.
The term ‘killer application’ was applied to Visicalc before Lotus existed. Without Visicalc there would never have been an IBM PC, middle managers would never have bought a word processor for their own use. Typing was for secretaries.
Lotus only got in the game because Kapor had written a graphing add-on to Visicalc on the PET/Apple/Tandy platforms of the day. Visicorp did not see the value of producing a version for the IBM PC, Kapor could not sell his add-on. So he wrote a combined spreadsheet/graphics program for the IBM PC and ate Visicorp’s lunch, dinner and breakfast.
And ten years or so later, Lotus gave the spreadsheet business away to Microsoft when they were too stupid to produce a version for either Windows or OS/2. The rivers of gold comming into Lotus could have easily paid for both.