![]() |
![]() |
#1 |
Bemusing Prompter
"Danny"
Dec 2002
California
2,467 Posts |
![]()
I've seen a few sources say that medical knowledge doubles every 73 days. It's interesting to hear that there is a Moore's law equivalent for health care. However, this doesn't seem to make sense for a few reasons.
For example, most people don't expect longevity escape velocity or a cure for Alzheimer's in 10 years. But if medical knowledge does double every 73 days, then it should double 50 times (a whole 15 orders of magnitude) within that timespan. Although human bodies are complicated things, I very highly doubt we have only a quadrillionth of the knowledge needed to cure aging or a disease like diabetes. Hell, there aren't even a quadrillion cells in our bodies! I've seen comments on Reddit saying that "knowledge" in this case means raw data rather actual knowledge that can be applied. This would explain why cures for many ailments are still years away. However, the amount of data still doesn't make sense as this would quickly overwhelm the world's total data capacity. We supposedly created 2,000 exabytes of medical data in 2020. If the data doubles every 73 days, then we would be looking at more than a decillion bytes of medical data in 2030 — much more than the projected 572 zettabytes of total data. And if medical data doubles every 73 days for 20 years, then the amount would increase by 30 orders of magnitude. We won't be able to store all that data even if we converted the entire Sun into hard drives! So is there something I'm missing? |
![]() |
![]() |
![]() |
#2 | |
"Serge"
Mar 2008
Phi(4,2^7658614+1)/2
100110100001102 Posts |
![]() Quote:
Suppose you have 2 Eb of MRI/CAT scans (or all genomes of the population of Iceland and/or England). What have you learned, simply from that? Not much. What have you learned from double that data? Double of not much. Mining this data is much like getting 2 grams of Pu from 2 million tons of ore. A lot of work still to be done. |
|
![]() |
![]() |
![]() |
#3 |
6809 > 6502
"""""""""""""""""""
Aug 2003
101×103 Posts
7×37×41 Posts |
![]()
That is like what is going to happen in astronomy when the LSST comes online at the Vera Rubin Observatory. They will be cranking out 20TB every night. The processing and delivery of that will be a challenge.
Last fiddled with by Uncwilly on 2022-05-08 at 22:57 |
![]() |
![]() |
![]() |
#4 |
Undefined
"The unspeakable one"
Jun 2006
My evil lair
2·3·5·7·31 Posts |
![]()
I think it simply boils down to a single statement:
Information != knowledge Having more data means only that, more data. You can't conflate that with understanding or insight. Although you might be able to conflate more data with more confusion. |
![]() |
![]() |
![]() |
#5 |
If I May
"Chris Halsall"
Sep 2002
Barbados
2×52×211 Posts |
![]() |
![]() |
![]() |
![]() |
#6 |
Bemusing Prompter
"Danny"
Dec 2002
California
2,467 Posts |
![]()
I figured "knowledge" meant raw data rather than actual knowledge in this case. However, this still doesn't make much sense. If data continues to double every 73 days, then we'll have 2100 times as much medical data as we do now in two decades. We can't store that amount of data even if the entire Sun was converted into data storage devices!
Last fiddled with by ixfd64 on 2022-05-09 at 00:50 |
![]() |
![]() |
![]() |
#7 | |
Undefined
"The unspeakable one"
Jun 2006
My evil lair
2×3×5×7×31 Posts |
![]() Quote:
But that number is only a number, nothing more. It doesn't tell us the future. You are reading too much into a single output. |
|
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Does it make sense to participate casually? | bur | Aliquot Sequences | 11 | 2021-05-03 15:01 |
Does this make sense? | Chuck | GPU to 72 | 2 | 2018-08-23 15:56 |
Does trial factoring to 75+ make sense | swl551 | Software | 10 | 2012-12-03 21:14 |
Knowledge | Zeta-Flux | Lounge | 9 | 2005-11-28 03:42 |
What is Randomness and does relative randomness make sense? | sakreha | Data | 1 | 2005-02-04 23:29 |