In the Gmail Genius vs Clueless Manager comparison, I think it's more apt to say that the genius can traverse the tree and 'drill down' where necessary, applying the most effective changes at the necessary level of detail. The Clueless Manager is always stuck at the highest levels of abstraction, and he can only work with very blunt objects. The point is that the largest gains can sometimes come from optimizing disk-seek speeds, and being able to recognize (and correct) this is an advantage over only being able to talk about performance in generalized terms.
I agree. Striking the right balance between abstractions and low-level details, is the mark of a genius. Put differently, the size and nature of the building blocks a person uses, in order to tackle a problem or to communicate with others is, to me, a very good indicator of how smart that person is.
I have been in situations where the balance has tipped far too much in either directions. Far too much reliance on abstractions, without understanding the details and we get the stock, garden variety Business person. Far too many details used in thinking and communicating leads to tedium, and interestingly can come off as patronizing, if its done often enough.
Yeah. Thinking too detailed too early on would be a mistake, too. Thinking about disk heads when you have no idea what the interface will look like is a huge mistake to me.
I don't think web developers have to think about disk I/O unless they're Google or Yahoo.
"... I don't think web developers have to think about disk I/O unless they're Google or Yahoo. ..."
I'm not sure about this.
For "any" interactive Webapp, speed is important. If you have repeated queries to/from, DB I/O does come into the picture when peak loads occur. So right when you get lots of users, the site will be hit. I've seen this with e-commerce (where user speed == $$$) where sure enough every peak time the server(s) would begin to grind. Remember your max speed will be determined by bandwidth. DB I/O will have some effect to slow response times. Maybe not as much as bandwidth but it is still something you must consider. Not every site is as extreme as e-commerce but the idea is still valid.
"... Every Web usability study I have conducted since 1994 has shown the same thing: users beg us to speed up page downloads ..." ~ http://www.useit.com/alertbox/9703a.html
Why is it important? Well user perception is one.
user: "Gee your application is slow compared to google, a bit slower than amazon"
developer: "But google invests squillions in hardware, network connections . Their applications run in memory. My apps run on fast hardware, I optomise for small pages ...."
user: "... So, it still appears slow"
If a user thinks your site is slow (and if you subscribe to the idea that you are creating web applications as apposed to documents) you need to keep your response times as close as possible to applications that people use every day.
It doesn't help that developers constantly throwing new (unoptomised) techniques to do new things at the same time larger sites are optomising to improve their load times, making comparisons between load times inevitable ~ http://developer.yahoo.net/blog/archives/performance/
Street fighters tend to be very good at details (street fighters as in people who never had formal training but kick ass). While dojo masters tend to be good in abstractions (not necessarily the useful kind). I think a good definition of a hacker is one who is good at both. I came from a dojo and got my ass kicked several times by a few street fighters. Since then I am learning the art of abstractions for details sake and not for its own sake.
Details matter, but when he says the smartest people disdain abstractions, I think he's off. The smartest people talk about certain selected details a lot. Other details are appropriately abstracted.
I don't think abstraction is even the right concept to be talking about here. Being able to work with an abstraction doesn't imply ignorance of the details, and ultimately the best code comes from people who know how to make abstractions that are elegant and aren't leaky (which takes expertise in both domains).
I'm not sure I agree with Aaron's implicit definition of genius (one who has a deep understanding of the details of things). I prefer to define genius as one who fundamentally changes the way we think about the field in which is genius operates.
By this definition, a genius can have an incredible talent for a field without having deconstructed the way their talent comes into being. They work on intuition. It is up to the people the follow them to figure out how they did it.
The system was down, costing the owner $10,000 per hour. No one could get it back up.
A "genius" was consulted, and after snooping around for a few minutes, logged in, and changed one byte. Everything worked perfectly.
"Thank you, thank you, thank you!" said the owner, "Give me a bill."
The bill was $10,000.
"I can't pay this bill until it's itemized," said the owner."
No problem -
Changing the byte: $1.
Knowing which byte to change: $9999.
Genius is:
a) Understanding that focus MUST be directed exactly where it's needed,
b) Being willing to do it, and
c) Being able to do it no matter where it is.
In the early years of this century, Steinmetz was brought to General
Electric's facilities in Schenectady, New York. GE had encountered a
performance problem with one of their huge electrical generators and had
been absolutely unable to correct it. Steinmetz, a genius in his
understanding of electromagnetic phenomena, was brought in as a consultant
-- not a very common occurrence in those days, as it would be now.
Steinmetz also found the problem difficult to diagnose, but for some days
he closeted himself with the generator, its engineering drawings, paper and
pencil. At the end of this period, he emerged, confident that he knew how
to correct the problem.
After he departed, GE's engineers found a large "X" marked with chalk on
the side of the generator casing. There also was a note instructing them to
cut the casing open at that location and remove so many turns of wire from
the stator. The generator would then function properly.
And indeed it did.
Steinmetz was asked what his fee would be. Having no idea in the world what
was appropriate, he replied with the absolutely unheard of answer that his
fee was $1000.
Stunned, the GE bureaucracy then required him to submit a formally itemized
invoice.
They soon received it. It included two items:
1. Marking chalk "X" on side of generator: $1.
2. Knowing where to mark chalk "X": $999.
Same for real artists. It's always the details which make the difference. And the reason is very clear: the details are the really delicate part (and the most difficult to grasp).
For someone, all sunsets are equal, and for some other one, every single one is a unique miracle.
