[18] Premature optimization is the root of all evil は、プログラミングに関する Donald Knuth の格言です >>19。
[7] Today's optimization is tomorrow's bottleneck (今日の最適化は明日のボトルネック) は、 安易なその場しのぎ的対処を戒め広い視野からの本質的な改善を良しとする格言です。
[16] 有名な言葉かと思っていたのだけど、 ネットワーク設計の分野を中心に広まったローカル用語なのかな。
[17] 誰が言い出したのかあんまよくわかんないし20世紀まで遡れるのかどうか微妙なラインにあるなあ。
Designing a network that is intelligently tuned (optimized) for a particular type of data or service — such as TV or financial transactions — inevitably makes that network less open. As software engineers say, "Today's optimization is tomorrow's bottleneck." Thus, the best network is a "stupid" network that does nothing but move bits.2
People
frequently call for optimization, for reasons including performance
needs, lack of reliability, security issues, and poor control over
bandwidth or latency. On the other hand, today's optimization is
tomorrow's roadblock; design choices made to optimize a particular
application may or may not prove the right ones to make when a new
application emerges. And optimization in a decentralized network such
as the Internet is delimited by difficulties in reaching agreement to
deploy optimizations network-wide and could lead to vendor lock-in.
There is a tension here between immediate improvements and long-term benefits: today's optimization may be tomorrow's roadblock, and design choices made to optimize a particular application may or may not turn out to be benefi-
[1] isen.blog: Today's Optimization, Tomorrow's Bottleneck () <http://isen.com/blog/2003/09/todays-optimization-tomorrows.html>
“Designing a network that is intelligently tuned (optimized)
for a particular type of data or service — such as TV or
financial transactions — inevitably makes that network less
open. As software engineers say, "Today's optimization is
tomorrow's bottleneck." “
David Isenberg and David Weinberger
Today's optimization is tomorrow's slug...
On the other hand, today's optimization is tomorrow's roadblock; design choices made to optimize a particular application may or may not prove the right ones to make when a new application emerges. And optimization in a decentralized network such as the Internet is delimited by difficulties in reaching agreement to deploy optimizations network-wide and could lead to vendor lock-in.
[6] QoS-transit services: end-to-end quality of service control in the Internet using dynamic pricing () <http://dspace.library.uvic.ca/handle/1828/2177>
今日の最適化は明日の障害
Today's optimization solution in product design for a technical problem will become tomorrow's performance bottleneck in the product.
[11] What optimizations today are going to be useless tomorrow? (2008, ) <http://www.stackprinter.com/export?service=stackoverflow&question=296297&printer=false&linktohome=true>
Today’s optimization is tomorrow’s
bottleneck
Grassmuck touched on aspects of the digital age, the 40th birthday of the ARPANET and the 10th birthday of Napster this year, and the end-to-end principles in system design laid out by Saltzer, Reed and Clark at the MIT. Laboratory for Computer Science in 1984. The network is a universal transport mechanism, intelligence and innovation are at the ends of the network. “Today’s optimization is tomorrow’s bottleneck.”
I also disagree that today's optimization is tomorrow's spam.