11 ms·

Chan Zuckerberg Initiative Makes Its First Acquisition, Meta

90 points by spuiszis 10y ago

divbit 10y ago
Facebook have produced a surprising amount of powerful tools recently, so getting into the research (as in research tools) space seems like it should be good for all (and obviously if they are deeply into AI research, having a good indexing tool is crucial).
- fezz 10y ago
  Some similarities to Kodak maybe.
Negative1 10y ago
Would it be accurate to say that Meta is essentially a social network for researchers and scientists?
igravious 10y ago
AI for Science, but not for the Arts & Humanities. With that slight snark I applaud this philanthropic move. :/ Tried out Meta once but couldn't get it to surface similar material to a set that I fed it. I don't need Meta to recommend by citations, I need it to recommend by topical similarity. Still though, props to team Zuck and I can see Meta complementing G Scholar or possibly supplanting it.
- 1_listerine_pls 10y ago
  > this philanthropic move?
- davidascher 10y ago
  CZI is focused on disease. A focus on biomed research is reasonable...
cing 10y ago
"All of Biohub's findings -- along with everything that is studied in the Chan Zuckerberg Science initiative -- will be open source and available to all." [1] I wonder if this extends to Meta? [1] https://www.engadget.com/2016/09/21/chan-zuckerberg-initiative-to-invest-3-billion-to-cure-disease/ https://www.engadget.com/2016/09/21/chan-zuckerberg-initiati...
- wavefunction 10y ago
  The article mentions "free to researchers", so apparently not all.
  - cing 10y ago
    Yeah, even if it's just for researchers, I interpreted that as "free to search", but not that the platform or data (citation graph w/ metadata) would be open source.
tabeth 10y ago
I've never heard of Meta, so, like any normal human being I went on their website. [1] TLDR: From a glance, this entire organization's existence is centered around the Chan Zuckerberg Initiative. There's no indication of anything useful they've done for real. I'm sure I could find that information if I bothered to look more, but for an organization that describes themselves as "mission-driven", and one that is "working on problems that matter," I fail to see any of those problems, or how this organization has contributed to the solution. I mean, not even a link to "Solutions" or "Mission" or "About". Nada. I'm very curious to how this page looked [yesterday]. Organizations that actually do stuff don't hesitate to put it front and center of their website, for example: - https://www.change.org/ https://www.change.org/ - https://www.planet.com/ https://www.planet.com/ This is literally a sentence by sentence breakdown of everything on Meta.com at the time of this posting. > "Most scientific breakthroughs have been preceded by the invention of new tools that help us see and experiment in new ways." Great! Oh, this is just a quote from Mark Zuckerberg, linking to the Chan Zuckerberg Initiative. OK. Moving along. > Meta is joining the Chan Zuckerberg Initiative. OK, got it. There's a bunch of corporate speak that you'll see if you click it. I won't click it because I want to know what Meta does outside of the whole CZI thing. > Reserve a free account > Meta is a tool that helps researchers understand what is happening globally in science and shows them where science is headed. Pending shareholder and court approval, the Chan Zuckerberg Initiative is acquiring Meta to help bring its technologies to the entire scientific community. Sign up now to reserve your free account. Neat! If you go to https://meta.science/reserve https://meta.science/reserve, where you actually can reserve you'll see more CZI stuff. Again, there's no indication of what these people do, in practice. > Join a mission-driven organization working on problems that matter Awesome! Oh, wait, this is just a link to their hiring page. What exactly do these people do, specifically? Sure they "help researchers understand what is happening globally", but that's so vague. Google and Wikipedia technically do that too in a way. [1] http://meta.com/ http://meta.com/
- cing 10y ago
  I've had an account at Meta prior to this announcement. The statement "Meta is a tool that helps researchers understand what is happening globally in science and shows them where science is headed" is not wrong, even though the landing page could use some work. What that amounts to for a researcher is a paper discovery service. You can search for papers/labs, and click on topics like "CRISPR/Cas9" and get the top 20 most impactful papers in the last decade using that technique. You can subscribe to new papers on that topic across a large set of journals. You can also observe trends in science, like certain research concepts that are increasing in popularity. In practice, as a researcher, I don't find it more useful than journal RSS feeds and a Google Scholar alert or two, but it's a step in the right direction.
dhairya 10y ago
Also check out Semantic Scholar. It's a publicly available tool from the Allen Institute for Artificial Intelligence. Currently supports CS research and neuroscience research, but we're working on expanding the domains in the upcoming years. One of the cool features is citation influence. You can see which citations are actually influential in an article. https://www.semanticscholar.org/ https://www.semanticscholar.org/
- loopasam 10y ago
  Also check out http://pubmed-watcher.org/ http://pubmed-watcher.org/ for biomedical domain
derpenator 10y ago
This company is based in toronto. A couple of months ago they fired everyone because they ran out of funding. They probably got bought out for cheap. It's got a really nice interface though, it's good that facebook will do something with it and for free, since their business plan sucked.
achiachi 10y ago
I'm just here because I thought this had to do with billionaire funded neocolonialism in Hawaii.
skosuri 10y ago
Hmm, I hope they got a good price. This doesn't really seem like a pain point for researchers (at least not for me). We are all pretty adept at using existing tools to find the papers we need to find. The bigger issue is time to properly read them. That said, I like the idea of CZI acting as a hub for tool development that will enable all researchers broadly. There are a bunch of tools that are incredibly useful to making logistics of a lab simpler. 1. Lab management and logistics: quartzy and others 2. Online paper writing: overleaf and a few others 3. DNA Bashing: benchling et al 4. Bioinformatics: They could do a bunch here I worry that a lot of things in this space right now will disappear pretty quickly once the VC gravy train runs dry. More importantly, I think this is something that CZI might have expertise to do well and sustainably.
- pseud0r 10y ago
  I don't know how meta works, but currently there is no quick way to do queries like: Find all studies which test compound X in concentrations between A and B sorted by organism, much less tools that use machine learning to extract the relevant information from thousands of papers and make simple summaries. What would be much more useful that making logistics of labs simpler is cheaper and more flexible lab automation to make all tasks in a lab programmable and automated.
  - skosuri 10y ago
    Pt1. Again, not sure this is useful, and has been tried many times with no useful end. The major issue is that simple summaries assume papers to be true. Just aggregating papers and making summaries is basically like search and reading abstracts. Understanding the true differences when within the morass is the hard part and usually is often very subtle. Pt2. A lot of people talk about automation, but biology has changed so fast in the last decade that any serious automation efforts in this space have become quickly outdated. It's useful when you are doing the same thing a million times, but such things are often already automated.
    - pseud0r 10y ago
      I don't disagree with that, but I do think there's massive improvement waiting to be done on scientific search engines. Not sure if there will be any big breakthrough until computer can read and understand the content of scientific papers to a much higher extent than today though. As for automation, I think that flexible automation solution shouldn't become outdated so fast. I think automating a lab to a large extent would be the same as automating a kitchen. You need to take thing in and out of storage, in a lab this would usually be different types of freezers, refrigerators, incubators etc., then you do stuff like mixing, slicing, dicing, heating, cooling, shaking, transferring liquids and solids between different types of equipment. Smaller labs often don't have much automation. There's plenty of labs that don't even have lab robots, even though they do plenty of pipetting. Automated storage solutions are also rare as far as I know. Just deploying existing solutions and making them cheaper would help. The only way to make thing really programmable though, is to automate everything. One challenge is that most machines and equipment only have interfaces for humans. If you were to place all your equipment in a kind of rack you could have a mix of conveyor belt for moving materials and robot arms on tracks in the roof to do manipulation tasks. Still it would be challenging to operate much of the equipment unless it is rebuilt to have standardized interfaces, but perhaps improvement in computer vision etc could eventually mitigate this. Another challenge is to make lab equipment communicate and compare data in completely different formats etc. There's lots of challenges but building a lab robotics system that can completely replace humans in doing the lab work should be possible even without hard AI and the hardware wouldn't necessarily have to be extremely expensive either.
kriro 10y ago
I'm troubled by the fact that they cannot open the tool immediately and that it will take "some time". I also don't see the value proposition over say Google Scholar+alerts or Researchgate trends. Wait and see I guess. The current website doesn't have much information other than the fact that Meta got bought and some generic blurbs. I'd much rather see Zuckerberg buy out an entire niche of science journals and Open Access them and keep them that way while keeping the high standards and impact factor in tact. Bioinformatics would have been a decent fit. I think freeing domains one by one with wads of tech-donor-$$$ might be the easiest stept towards open science.
bamboozled 10y ago
Here is another "Chan Zuckerberg" initiative I'm sure everyone will be proud of: Hawaiians call Mark Zuckerberg 'the face of neocolonialism' over land lawsuits It seems greed knows no bounds. Throwing money at things is not charity. [1] https://www.theguardian.com/technology/2017/jan/23/mark-zuckerberg-hawaii-land-lawsuits-kauai-estate https://www.theguardian.com/technology/2017/jan/23/mark-zuck...
- khazhou 10y ago
  Interesting point. Like you, I'm also skeptical of the ability of this tool to accurately detect useful research claims/results, and surface them to the right user. It's easy these days to claim "Deep learning" and "Machine intelligence" and get significant funding. If MarkZ helped oversee the due diligence of this company (and I don't know how involved he is, since this is Chan's initiative) then they probably were evaluated very thoroughly, given Mark's recent interest in AI.
- l33tbro 10y ago
  Nobody lost their land to a bloodthirsty colonialist. These indigenous families opted to sell their land and were compensated at some point. So please elaborate on what you mean by Zuckerberg's boundless "greed"? Were I indigenous, I think I'd actually find this kind of Western academic white-knighting slightly offensive. You'd be essentially questioning the agency, sovereignty, and decision-making capacity of my people's ability manage resources. The partial land-owners with lineage to the original 14 parcels of land on the North Island are being compensated for something then never knew they had, and are not being evicted from the land. Those that sold their land also initially knew what they were doing. So it's unfair to hold Zuckerberg solely accountable for the market forces that dispossess the indigenous of their land.
  - bamboozled 10y ago
    If you do some more research you will find that some people will indeed be forced off their land if he gets his way. They will be compensated but they will be forced off their land. There is a major difference as these people aren't interested in revenue, some of them haven't even been made aware they need to appear in court. The following write up explains the issue in more detail and there are countless more articles and blog posts explaining the situation http://www.huffingtonpost.com/entry/mark-zuckerberg-kauai-lawsuits_us_588113a6e4b0e3a73567bc45 http://www.huffingtonpost.com/entry/mark-zuckerberg-kauai-la... Mark is greedy because it's just a holiday house that he doesn't really need to own. He is just causing a bunch of people grief so he can have his own way and his own playground. Ironically he wants privacy (haha). He is bullying people with lawyers and money. If he wants to be Gordon Gecko type figure, be Gordon Gecko. But this generous philanthropist angle is a hoax when he is behaving like this.

90 points by spuiszis 10y ago

Facebook have produced a surprising amount of powerful tools recently, so getting into the research (as in research tools) space seems like it should be good for all (and obviously if they are deeply into AI research, having a good indexing tool is crucial).

fezz 10y ago
Some similarities to Kodak maybe.

Negative1 10y ago

Would it be accurate to say that Meta is essentially a social network for researchers and scientists?

igravious 10y ago

AI for Science, but not for the Arts & Humanities. With that slight snark I applaud this philanthropic move. :/ Tried out Meta once but couldn't get it to surface similar material to a set that I fed it. I don't need Meta to recommend by citations, I need it to recommend by topical similarity. Still though, props to team Zuck and I can see Meta complementing G Scholar or possibly supplanting it.

1_listerine_pls 10y ago
> this philanthropic move?
davidascher 10y ago
CZI is focused on disease. A focus on biomed research is reasonable...

cing 10y ago

"All of Biohub's findings -- along with everything that is studied in the Chan Zuckerberg Science initiative -- will be open source and available to all." [1] I wonder if this extends to Meta? [1] https://www.engadget.com/2016/09/21/chan-zuckerberg-initiative-to-invest-3-billion-to-cure-disease/ https://www.engadget.com/2016/09/21/chan-zuckerberg-initiati...

wavefunction 10y ago
The article mentions "free to researchers", so apparently not all.
- cing 10y ago
  Yeah, even if it's just for researchers, I interpreted that as "free to search", but not that the platform or data (citation graph w/ metadata) would be open source.

tabeth 10y ago

I've never heard of Meta, so, like any normal human being I went on their website. [1] TLDR: From a glance, this entire organization's existence is centered around the Chan Zuckerberg Initiative. There's no indication of anything useful they've done for real. I'm sure I could find that information if I bothered to look more, but for an organization that describes themselves as "mission-driven", and one that is "working on problems that matter," I fail to see any of those problems, or how this organization has contributed to the solution. I mean, not even a link to "Solutions" or "Mission" or "About". Nada. I'm very curious to how this page looked [yesterday]. Organizations that actually do stuff don't hesitate to put it front and center of their website, for example: - https://www.change.org/ https://www.change.org/ - https://www.planet.com/ https://www.planet.com/ This is literally a sentence by sentence breakdown of everything on Meta.com at the time of this posting. > "Most scientific breakthroughs have been preceded by the invention of new tools that help us see and experiment in new ways." Great! Oh, this is just a quote from Mark Zuckerberg, linking to the Chan Zuckerberg Initiative. OK. Moving along. > Meta is joining the Chan Zuckerberg Initiative. OK, got it. There's a bunch of corporate speak that you'll see if you click it. I won't click it because I want to know what Meta does outside of the whole CZI thing. > Reserve a free account > Meta is a tool that helps researchers understand what is happening globally in science and shows them where science is headed. Pending shareholder and court approval, the Chan Zuckerberg Initiative is acquiring Meta to help bring its technologies to the entire scientific community. Sign up now to reserve your free account. Neat! If you go to https://meta.science/reserve https://meta.science/reserve, where you actually can reserve you'll see more CZI stuff. Again, there's no indication of what these people do, in practice. > Join a mission-driven organization working on problems that matter Awesome! Oh, wait, this is just a link to their hiring page. What exactly do these people do, specifically? Sure they "help researchers understand what is happening globally", but that's so vague. Google and Wikipedia technically do that too in a way. [1] http://meta.com/ http://meta.com/

cing 10y ago
I've had an account at Meta prior to this announcement. The statement "Meta is a tool that helps researchers understand what is happening globally in science and shows them where science is headed" is not wrong, even though the landing page could use some work. What that amounts to for a researcher is a paper discovery service. You can search for papers/labs, and click on topics like "CRISPR/Cas9" and get the top 20 most impactful papers in the last decade using that technique. You can subscribe to new papers on that topic across a large set of journals. You can also observe trends in science, like certain research concepts that are increasing in popularity. In practice, as a researcher, I don't find it more useful than journal RSS feeds and a Google Scholar alert or two, but it's a step in the right direction.

dhairya 10y ago

Also check out Semantic Scholar. It's a publicly available tool from the Allen Institute for Artificial Intelligence. Currently supports CS research and neuroscience research, but we're working on expanding the domains in the upcoming years. One of the cool features is citation influence. You can see which citations are actually influential in an article. https://www.semanticscholar.org/ https://www.semanticscholar.org/

loopasam 10y ago
Also check out http://pubmed-watcher.org/ http://pubmed-watcher.org/ for biomedical domain

derpenator 10y ago

This company is based in toronto. A couple of months ago they fired everyone because they ran out of funding. They probably got bought out for cheap. It's got a really nice interface though, it's good that facebook will do something with it and for free, since their business plan sucked.

achiachi 10y ago

I'm just here because I thought this had to do with billionaire funded neocolonialism in Hawaii.

skosuri 10y ago

Hmm, I hope they got a good price. This doesn't really seem like a pain point for researchers (at least not for me). We are all pretty adept at using existing tools to find the papers we need to find. The bigger issue is time to properly read them. That said, I like the idea of CZI acting as a hub for tool development that will enable all researchers broadly. There are a bunch of tools that are incredibly useful to making logistics of a lab simpler. 1. Lab management and logistics: quartzy and others 2. Online paper writing: overleaf and a few others 3. DNA Bashing: benchling et al 4. Bioinformatics: They could do a bunch here I worry that a lot of things in this space right now will disappear pretty quickly once the VC gravy train runs dry. More importantly, I think this is something that CZI might have expertise to do well and sustainably.

pseud0r 10y ago
I don't know how meta works, but currently there is no quick way to do queries like: Find all studies which test compound X in concentrations between A and B sorted by organism, much less tools that use machine learning to extract the relevant information from thousands of papers and make simple summaries. What would be much more useful that making logistics of labs simpler is cheaper and more flexible lab automation to make all tasks in a lab programmable and automated.
- skosuri 10y ago
  Pt1. Again, not sure this is useful, and has been tried many times with no useful end. The major issue is that simple summaries assume papers to be true. Just aggregating papers and making summaries is basically like search and reading abstracts. Understanding the true differences when within the morass is the hard part and usually is often very subtle. Pt2. A lot of people talk about automation, but biology has changed so fast in the last decade that any serious automation efforts in this space have become quickly outdated. It's useful when you are doing the same thing a million times, but such things are often already automated.
  - pseud0r 10y ago
    I don't disagree with that, but I do think there's massive improvement waiting to be done on scientific search engines. Not sure if there will be any big breakthrough until computer can read and understand the content of scientific papers to a much higher extent than today though. As for automation, I think that flexible automation solution shouldn't become outdated so fast. I think automating a lab to a large extent would be the same as automating a kitchen. You need to take thing in and out of storage, in a lab this would usually be different types of freezers, refrigerators, incubators etc., then you do stuff like mixing, slicing, dicing, heating, cooling, shaking, transferring liquids and solids between different types of equipment. Smaller labs often don't have much automation. There's plenty of labs that don't even have lab robots, even though they do plenty of pipetting. Automated storage solutions are also rare as far as I know. Just deploying existing solutions and making them cheaper would help. The only way to make thing really programmable though, is to automate everything. One challenge is that most machines and equipment only have interfaces for humans. If you were to place all your equipment in a kind of rack you could have a mix of conveyor belt for moving materials and robot arms on tracks in the roof to do manipulation tasks. Still it would be challenging to operate much of the equipment unless it is rebuilt to have standardized interfaces, but perhaps improvement in computer vision etc could eventually mitigate this. Another challenge is to make lab equipment communicate and compare data in completely different formats etc. There's lots of challenges but building a lab robotics system that can completely replace humans in doing the lab work should be possible even without hard AI and the hardware wouldn't necessarily have to be extremely expensive either.

kriro 10y ago

I'm troubled by the fact that they cannot open the tool immediately and that it will take "some time". I also don't see the value proposition over say Google Scholar+alerts or Researchgate trends. Wait and see I guess. The current website doesn't have much information other than the fact that Meta got bought and some generic blurbs. I'd much rather see Zuckerberg buy out an entire niche of science journals and Open Access them and keep them that way while keeping the high standards and impact factor in tact. Bioinformatics would have been a decent fit. I think freeing domains one by one with wads of tech-donor-$$$ might be the easiest stept towards open science.

bamboozled 10y ago

Here is another "Chan Zuckerberg" initiative I'm sure everyone will be proud of: Hawaiians call Mark Zuckerberg 'the face of neocolonialism' over land lawsuits It seems greed knows no bounds. Throwing money at things is not charity. [1] https://www.theguardian.com/technology/2017/jan/23/mark-zuckerberg-hawaii-land-lawsuits-kauai-estate https://www.theguardian.com/technology/2017/jan/23/mark-zuck...

khazhou 10y ago
Interesting point. Like you, I'm also skeptical of the ability of this tool to accurately detect useful research claims/results, and surface them to the right user. It's easy these days to claim "Deep learning" and "Machine intelligence" and get significant funding. If MarkZ helped oversee the due diligence of this company (and I don't know how involved he is, since this is Chan's initiative) then they probably were evaluated very thoroughly, given Mark's recent interest in AI.
l33tbro 10y ago
Nobody lost their land to a bloodthirsty colonialist. These indigenous families opted to sell their land and were compensated at some point. So please elaborate on what you mean by Zuckerberg's boundless "greed"? Were I indigenous, I think I'd actually find this kind of Western academic white-knighting slightly offensive. You'd be essentially questioning the agency, sovereignty, and decision-making capacity of my people's ability manage resources. The partial land-owners with lineage to the original 14 parcels of land on the North Island are being compensated for something then never knew they had, and are not being evicted from the land. Those that sold their land also initially knew what they were doing. So it's unfair to hold Zuckerberg solely accountable for the market forces that dispossess the indigenous of their land.
- bamboozled 10y ago
  If you do some more research you will find that some people will indeed be forced off their land if he gets his way. They will be compensated but they will be forced off their land. There is a major difference as these people aren't interested in revenue, some of them haven't even been made aware they need to appear in court. The following write up explains the issue in more detail and there are countless more articles and blog posts explaining the situation http://www.huffingtonpost.com/entry/mark-zuckerberg-kauai-lawsuits_us_588113a6e4b0e3a73567bc45 http://www.huffingtonpost.com/entry/mark-zuckerberg-kauai-la... Mark is greedy because it's just a holiday house that he doesn't really need to own. He is just causing a bunch of people grief so he can have his own way and his own playground. Ironically he wants privacy (haha). He is bullying people with lawyers and money. If he wants to be Gordon Gecko type figure, be Gordon Gecko. But this generous philanthropist angle is a hoax when he is behaving like this.