4 ms·

This reminds me of the infamous comparisons between the M16 and the AK47 assault rifles. The M16 is a superior weapon by almost every standard - it's crafted ou

This reminds me of the infamous comparisons between the M16 and the AK47 assault rifles. The M16 is a superior weapon by almost every standard - it's crafted out of lighter, aeronautical materials, it has a beautiful, elegant design, it's far more precise and has far less recoil. Except nobody cares. What people want is to pull their weapon out of a muddy hole after it's been there for a week, and have it reliably shoot thirty rounds.

roc 17y ago
That's what some people want. Some of the time. Which I believe is the point of the article. Unless you know all the circumstances of what someone needs a weapon for, you can't say the M16 is better or worse than the AK47. (People who don't have to store their weapon in a muddy hole or go weeks without resupply may and do prefer the M16 for its characteristics) Simply: there is no universal measurement of quality for any designed product. Even hammers have trade-offs in their design that prevent absolute quality measurements.
- wensing 17y ago
  But you're already talking about two very fine guns. Just because you have to consider trade-offs at that level doesn't mean there's no qualitative scale at all, does it?
  - roc 17y ago
    Yes and No. Sure, there are guns objectively worse than both the M16 and AK47 in any context. But beyond trivial examples of almost implausibly bad knockoffs, a universal qualitative scale remains impossible to define. Perhaps it would seem less wild a concept if framed as: you can't build a truly useful qualitative scale until you know the context. You can trivially sort things into 'good' and 'crap' - it just won't get you very far.
- jerf 17y ago
  The concept of Pareto optimality is very useful here, though I'd rather suggest you hit Wikipedia than try to describe it. http://en.wikipedia.org/wiki/Pareto_efficiency http://en.wikipedia.org/wiki/Pareto_efficiency
- coffeemug 17y ago
  I disagree. That's what most people want most of the time. There are specialized situations where M16 is a better choice, but for the most part people want the AK47. The numbers speak for themselves - more AK47s were sold in the world than all other assault rifles combined. There is no universal scale of quality, but if you're trying to hit it big you have to keep in mind what most people want most of the time, and it's usually not what first comes to mind.
  - dagw 17y ago
    more AK47s were sold in the world than all other assault rifles combined. Don't you think that that is more due to price and availability than anything else. I imagine buying a case of M16s can be quite tricky in the middle of the mountains in North Pakistan, while you probably have at least two people in your circle of friends who knows I guy who makes AK47s in his workshop. So to draw the AK47 analogy to its logical conclusion, If you want your product to hit it big make it easy to build give away plans on how to built it to everybody without any sort patent, copyright or royalty claims. Remember that the vast majority or AK47s bought are knockoffs and that the original company doesn't see a dime from those sales.

roc 17y ago

That's what some people want. Some of the time. Which I believe is the point of the article. Unless you know all the circumstances of what someone needs a weapon for, you can't say the M16 is better or worse than the AK47. (People who don't have to store their weapon in a muddy hole or go weeks without resupply may and do prefer the M16 for its characteristics) Simply: there is no universal measurement of quality for any designed product. Even hammers have trade-offs in their design that prevent absolute quality measurements.

wensing 17y ago
But you're already talking about two very fine guns. Just because you have to consider trade-offs at that level doesn't mean there's no qualitative scale at all, does it?
- roc 17y ago
  Yes and No. Sure, there are guns objectively worse than both the M16 and AK47 in any context. But beyond trivial examples of almost implausibly bad knockoffs, a universal qualitative scale remains impossible to define. Perhaps it would seem less wild a concept if framed as: you can't build a truly useful qualitative scale until you know the context. You can trivially sort things into 'good' and 'crap' - it just won't get you very far.
jerf 17y ago
The concept of Pareto optimality is very useful here, though I'd rather suggest you hit Wikipedia than try to describe it. http://en.wikipedia.org/wiki/Pareto_efficiency http://en.wikipedia.org/wiki/Pareto_efficiency
coffeemug 17y ago
I disagree. That's what most people want most of the time. There are specialized situations where M16 is a better choice, but for the most part people want the AK47. The numbers speak for themselves - more AK47s were sold in the world than all other assault rifles combined. There is no universal scale of quality, but if you're trying to hit it big you have to keep in mind what most people want most of the time, and it's usually not what first comes to mind.
- dagw 17y ago
  more AK47s were sold in the world than all other assault rifles combined. Don't you think that that is more due to price and availability than anything else. I imagine buying a case of M16s can be quite tricky in the middle of the mountains in North Pakistan, while you probably have at least two people in your circle of friends who knows I guy who makes AK47s in his workshop. So to draw the AK47 analogy to its logical conclusion, If you want your product to hit it big make it easy to build give away plans on how to built it to everybody without any sort patent, copyright or royalty claims. Remember that the vast majority or AK47s bought are knockoffs and that the original company doesn't see a dime from those sales.