The more things change, the more they remain the same – so seems to be the case with Steve Jobs & his approach to new products – build a simple, but robust core for the first iteration & then develop it over time. It worked for the Mac, then iPhone & seems to be working for the iPad. The article also touches on the closed Apple ecosystem & Jobs hostility towards flash (apparently similar to the arrow keys on the original Mac)
An interesting take on Apple & Steve Jobs closed approach
Quite a collection ranging from mainstream Windows screenshots to oldies like Xerox Alto
I’ve been doing quite a bit of travelling in taxis in Mumbai over the last 4-5 months, and one of the interesting things I noticed that the fares on the taxis with mechanical meters usually comes out higher than taxis with digital ones. This is especially true when there is a significant amount of waiting time due to heavy traffic. Here’s my theory on why this is so:
Faulty waiting time calibration on mechanical meters
On mechanical meters, we use a reference chart to convert the meter reading to the appropriate fare. All this was fine when the meters came out originally many years back and the meters were calibrated for a particular waiting fare rate. However, after several fare hikes that just raised the rate per km & not the waiting time rates, this calibration has become erroneous.
An example is in order to explain this. Say, initially 1.00 on the meter meant Rs 10 (per km) and 0.10 on the meter corresponded to 2 min waiting time at the rate of Rs 0.50/min (effectively meaning that for every 0.10 you pay Re 1, i.e., the same as the per km rate). Here, we have a uniform multiplication factor of 10 for both distance and waiting time.
Now, let’s say that there have a series of revisions and the rate per km has doubled, but the waiting rate is still the same. So, we should have 1.00 on the meter corresponding to Rs 20 (per km) as the distance rate, while 0.10 still signifies a 2 min waiting time at the rate of Rs 0.50/min. Therefore, the multiplication factors are now different for distance (20) and waiting time (still 10).
However, the fare charts are created only keeping the distance fares in mind, due to which you have the following scenario: For a trip of 2 km with a waiting time of 10 mins, the meter will read 2.50 (2×1.00 + 10×0.50×0.10) for both old and new rates.
- As per the initial rates, the fare would be Rs 25 (meter: 2.50×10 or rate breakup: 2×10 + 10×0.50)
- For the new rates, the actual fare should be Rs 45 (2×20 + 10×0.50)
- However, the new rate chart prepared would have only factored the increase in per km rates and would suggest a uniform multiplication factor of 20 for the meter reading, due to which you would end up paying Rs 50 (2.50×20)
Long live digital meters?
In the case of digital meters, they are recalibrated (at least in Mumbai, but not so much in Kolkata due to which the same problem exists) for the new fares without changing the waiting rates. Due to this you end up paying the actual fare (Rs 45 from the example above) when you use a taxi with a digital meter.
Of course, if the driver forgets to wind his mechanical meter before your trip, you end up avoiding the waiting charges altogether which gives you the lowest possible fare. So, I guess there’s a flip side to the whole mechanical vs. digital meter argument.
A budget simulation game (US centric)
In the old days (before the Internet), no technology products were free, because distribution costs made it impossible to offer anything without some commitment from the end customer. As a result, new technology adoption generally started with the deepest pockets (the military) and worked its way down to the shallowest pockets (the consumer). Since the introduction of the Internet, many technology products can be distributed for free, and therefore have some free or free trial version. Interestingly, the order of adoption now follows decision-making speed rather than deep pockets. That is, consumers who can decide very quickly adopt first and the military — who has a notoriously complex decision making process — adopts last.
Highlights the good old GIGO principle for computers, and when all else fails, "Heads or tails, gentlemen?"
Nicely summarises the importance of shipping the first version of your product using Apple as an example. Apple does sacrifice features in its products very often. The mantra seems to be "usage is like oxygen for ideas"
Kind of like the Chinese whisper phenomenon
Includes some gems of laws:
Weinberg’s Law of Metrics: “That which gets measured gets fudged.”
The Metric Law of 90s: “The first 90 percent of a development project takes 90 percent of the schedule. The remaining 10 percent of the project takes the other 90 percent of the schedule.”
The Metric Law of Least Resistance: “The more human effort required to calculate a metric, the less often (and less accurately) it will be calculated, until it is abandoned or ignored altogether.”
Bruce F. Webster's personal experience on how well a Cravath system could work, & some possible pitfalls on the road to implementing such a system.
Pretty radical thoughts based on the Cravath System: "Bring lots of new employees in, team them up with mentors, provide real work to do, and give them a choice: either get lots of great experience and get out, or work hard for a higher-up position."