So much talk about data, but no funder wants to support databases.
-
So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
This is a problem that people have been warning about for almost 15 years now, see e.g.:
Slide 13-17 from 2012.
How can it be that so often in science, most people understand and agree that there is a massive problem and yet, nothing happens.
When the Flybase shit hit the fan, I contacted them and asked them for their plans for decentralization - they had none, they said.
-
This is a problem that people have been warning about for almost 15 years now, see e.g.:
Slide 13-17 from 2012.
How can it be that so often in science, most people understand and agree that there is a massive problem and yet, nothing happens.
When the Flybase shit hit the fan, I contacted them and asked them for their plans for decentralization - they had none, they said.
I'm at a loss for words.This is a recurring topic in science and I still don't understand how this can be so common.
-
I'm at a loss for words.This is a recurring topic in science and I still don't understand how this can be so common.
Point also being that no scientist has any training on how to organise datasets into useful databases, and the coordination overheads (not to speak of the credit-sharing issues) aren't ignorable when setting up a decentralized system.
-
Point also being that no scientist has any training on how to organise datasets into useful databases, and the coordination overheads (not to speak of the credit-sharing issues) aren't ignorable when setting up a decentralized system.
Absolutely! This is not trivial, but imo needs to be done for all the ca. 1400 databases we have in all of biology.
Trivial or not, my guess would be that 15 years should be enough time?
-
I'm at a loss for words.This is a recurring topic in science and I still don't understand how this can be so common.
@brembs @albertcardona look at the incentive structures in science now. It’s all innovation, newness is fetishised, publish publish publish, only “high impact” journal publications really matter … where is the place for kind of boring, routine maintenance and enabling of others?
-
So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
@albertcardona Agree. Thousands of hours of stupid cat videos are produced & stored every minute but there's no provisioning for research data? Databrokers accrue (as of 2017) 2.7 Mb of data on every American every day, but we can't find the fucking storage for decades of empirical findings? It's a serious priorities alignment mismatch.
...Maybe the framing is wrong.
O_o
Just had a bit of inspiration whisper in my ear. Hear me out, doc. It may sound a little crazy, but it might just work! -
So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
@albertcardona Let us take advantage of someone elses distress. Insomch as it can be considered that given who i'm about to suggest. I just read a few days ago that Elon Musk was mothballing one of his big server farms. Now i'm no fan of the man, personally. But he *IS* easy to sway, with the right kind of argument & pitch, which this fits VERY neatly.
What if an appeal was made to host all that data at his fancy shmancy server farm he's about to let go fallow? It'd be a tax writeoff for him. -
So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
@albertcardona Now i'd stress the importance of contractual guarantees beforehand. IRONCLAD ones. The man is Lt Colonel of the ADHD brigade. The only thing that changes faster than his whimsey is how many women he's fathered children with. So it'd have to be moved to some sort of trust to have any guarantee of permanence. But for him, it would be a massive writeoff on an asset set to go fallow he'd still be paying taxes on. It's a win/win for both parties & FOR SCIENCE!<In best dexters lab riff>
-
@brembs @albertcardona look at the incentive structures in science now. It’s all innovation, newness is fetishised, publish publish publish, only “high impact” journal publications really matter … where is the place for kind of boring, routine maintenance and enabling of others?
That is correct, yes, but a) the people who run, e.g., Flybase now are, to my knowledge, not scientists in this aspect any more and b) if you're going to have a database at all, why not do it right?
-
So much talk about data, but no funder wants to support databases.
FlyBase is in trouble, yet a key reference in biology research.
Raising funds to support curating connectome datasets, including their large, nanometer-resolution volumes, is hard. Yet then evaluators point out that we aren't doing enough to make it easy to access – yet with what funds?
And likewise for every scientific dataset ... either there is an internally funded institute behind it (and even then) or it is hell.
Everyone wants new data, but nobody seems to pay attention to the need for old data to support the generation and interpretation of new data!
@albertcardona same story for emousealtas, an amazing invaluable resource built up over many years, then it limped along unfunded for 5 years before being decommisioned in 2023
https://www.emouseatlas.org/emap/home.html -
That is correct, yes, but a) the people who run, e.g., Flybase now are, to my knowledge, not scientists in this aspect any more and b) if you're going to have a database at all, why not do it right?
@brembs @albertcardona oh absolutely. My rant was about science/academia in general, not FlyBase specifically.
-
R relay@relay.infosec.exchange shared this topic
