11–20 of 68 posts
A big misconception about software among laypeople is it doesn’t need maintenance. It’s just not true.
Even the best of them need maintenance and up gradation for the simple reason that the environment in which operates is continuously changing.
Error bars to uncertainty of facts presented. Confidence. Sometimes I think the bandwidth/delay product is insufficiently appreciated but I can't quite say why. Db are widely misunderstood. And RMS. Without a reference, 2x bigger is meaningless. Likewise s/n ratio is misunderstood. Scale free can be OK, but usually not. People need to know that cells don't have walls and are not bags of liquid. At the scale we're tal…
v = dx/dt If you are driving 5 miles across town (dx), and you average 25 mph (v), it will take you roughly 5/25 hours, or 12 minutes (dt) to get there. By driving 50 miles per hour, you get there in 6 minutes. Is it worth the risk to yourself and others (bikers and pedestrians) to save 6 minutes? If you are driving to work and it normally takes you 30 minutes to get there by going 60 mph (30 miles), then if you go 8…
a = dv /dt.
The gist: consider a yes/no test that has a small chance of being wrong (false positive and false negative). Examples: spam filter, virus filter, pregnancy test, testing for another illness, etc.
How well it performs in reality depends enormously on the real ratio of yes vs. no.
Example:
Spam filter is 99% correct, 1% false positive and equally 1% false negative. That sounds good! And it is - IF your spam vs ham rate is 50-50. If, however, 99% of mails are spam, and you use this filter on 10,000 mails, then:
- 9900 are spam; 100 ham
- 1% of ham = 1 mail ends up in spamfolder
- 1% of spam = 99 mails end up in inbox.
So your inbox now contains 99 ham mails and 99 spam - an unacceptable 50-50 ratio, even though your using a superduper spam filter.
Changing the base rate of spam affects the outcome drastically. Obviously, 99% ham and only 1% spam flips the numbers around: inbox is fine, but 50% of your spamfolder isn't spam. Also not really okay -- and that with exactly the same filter, just a different base rate.
TL;DR: actual effectiveness of tests which have false positives/negatives is heavily skewed by the frequency of occurrence of whatever you're testing for.
v = dx/dt If you are driving 5 miles across town (dx), and you average 25 mph (v), it will take you roughly 5/25 hours, or 12 minutes (dt) to get there. By driving 50 miles per hour, you get there in 6 minutes. Is it worth the risk to yourself and others (bikers and pedestrians) to save 6 minutes? If you are driving to work and it normally takes you 30 minutes to get there by going 60 mph (30 miles), then if you go 8…
v = dx/dt If you are driving 5 miles across town (dx), and you average 25 mph (v), it will take you roughly 5/25 hours, or 12 minutes (dt) to get there. By driving 50 miles per hour, you get there in 6 minutes. Is it worth the risk to yourself and others (bikers and pedestrians) to save 6 minutes? If you are driving to work and it normally takes you 30 minutes to get there by going 60 mph (30 miles), then if you go 8…
And, by association, F = m*V^2 The risk of serious injury in an accident increases exponentially with speed. This is especially true for pedestrian victims.
Base Rate fallacy. The gist: consider a yes/no test that has a small chance of being wrong (false positive and false negative). Examples: spam filter, virus filter, pregnancy test, testing for another illness, etc. How well it performs in reality depends enormously on the real ratio of yes vs. no. Example: Spam filter is 99% correct, 1% false positive and equally 1% false negative. That sounds good! And it is - IF yo…
- Conditional probability
- Bayes' Theorem
v = dx/dt If you are driving 5 miles across town (dx), and you average 25 mph (v), it will take you roughly 5/25 hours, or 12 minutes (dt) to get there. By driving 50 miles per hour, you get there in 6 minutes. Is it worth the risk to yourself and others (bikers and pedestrians) to save 6 minutes? If you are driving to work and it normally takes you 30 minutes to get there by going 60 mph (30 miles), then if you go 8…
1.)Quantum entanglement, to instill a sense of appreciation for the unknown. 2.)Astrology, to grant an understanding of the language of archetypes.
Was that humour?
Error bars to uncertainty of facts presented. Confidence. Sometimes I think the bandwidth/delay product is insufficiently appreciated but I can't quite say why. Db are widely misunderstood. And RMS. Without a reference, 2x bigger is meaningless. Likewise s/n ratio is misunderstood. Scale free can be OK, but usually not. People need to know that cells don't have walls and are not bags of liquid. At the scale we're tal…