6 ms·

Getting started on Twitter as a scientist

96 points by amrrs 7y ago

Traster 7y ago
One thing that this article misses about twitter is that you carry your followers around with you. So if you have 1000 followers and you tweet at someone, whatever you're saying, you're saying in front of 1000 of your fans. So if you disagree with someone not only are they going to get your criticism, they're likely to get 1000 people piling on behind you. There are lots of things I can say to you one on one that are suddenly basically bullying if I've got a mob behind me. Equally, if you draw attention to someone, even if you aren't disagreeing with them, you have to be aware that you're likely subjecting them to a level of unwanted attention from your audience.
- bosswipe 7y ago
  If you reply to someone's tweet then mostly the people who follow both accounts will see the reply, in addition Twitter's algorithms will opaquely decide whether to show the reply to your other followers. If you want to ensure that your followers see the reply you can use Retweet with Comment.
azangru 7y ago
> There is no single guideline on when to block somebody—some may only support blocking trolls, and refrain from blocking other scientists who may be critical, while others believe it’s perfectly fine to mute or block any account... A good rule of thumb might be to block somebody if they have a repeated negative effect on your Twitter experience > It is perfectly okay to not look at Twitter for hours or even days. > Do not worry that this might be confusing: Twitter users can easily skip over tweets in languages they are not proficient in... Who is the target audience of a text like this? Do people feel they need someone's permission to block someone, or to skip Twitter for a while, or to write in whichever language they please? Honestly!
- thrwaway69 7y ago
  The point is to commoditize yourself so that's not unexpected. The brand (you) has to get good at targeting the audience they are selling product (yourself) to and avoid interacting with the competition or getting run over by trolls. Protecting the brand (yourself) is important. To distinguish the brand (yourself) from others, the brand (you) has to set something people can look up to as a way to differentiate their experience with them (you). What the brand (you) likes, follows, engages with etc defines the image of them (you) and the right market fit (their feed). Accessibility is a huge factor for a commodity (sellable parts of you as a whole) whilst some blocking may be necessary for setting the cultural expectations for the brand (you). If the brand (you) block people from certain group and it is public, that might work as a signal for what the brand (you) stands for. Let's suppose the brand (I) blocks a mark Zuckerberg and Facebook, then that works as a signal for that the brand (I) cares about privacy and against people who exploit that. If the brand (I) only uses English to tweet, then that might be a signal for which audience product (I) is made for. If the brands (yours) block list isn't diverse, it might signal as a racial/ethnic/gender/sex bias.
- im3w1l 7y ago
  It's not about "permission". It's about popularity. If you write too many posts in languages your followers don't speak, they might curate you out of their feeds decreasing your popularity. As a newcomer you don't know how important it is to take this into account.
- Mountain_Skies 7y ago
  The president was sued for blocking people on Twitter and ordered by a court to not block anymore. That might create some confusion especially for anyone making a Twitter account as an official communication medium for a government facility.
knzhou 7y ago
As a scientist I have a strong aversion to getting on Twitter, and this article isn't helping. All my experience with Twitter has told me it's a concentrator for hot (i.e. wrong) takes, snarky dismissals, and sloganeering, to a greater extent than any other social media platform. It seems to degrade thoughtful people into ordinary ones and ordinary people into monsters. I'll take even HN flamewars over that any day.
- SkyMarshal 7y ago
  >It seems to degrade thoughtful people into ordinary ones and ordinary people into monsters. In my experience that’s about 90% true, but there’s a small minority of people who remain thoughtful and insightful even in 280 characters. If you can find them and mutually follow each other, and then mute or block all the blowhards, tryhards and self-promoters, you’ll have cultivated a useful network of colleagues beyond what you might otherwise have via work and conferences. And if multi-disciplinary perspectives are valuable to your own research, then that Twitter network is especially valuable in that regard, as it’s likely to be at least as multidisplinary as your day-to-day work environment and more-so than your usual conference circuit. That said, it does take some work to cultivate that graph. May or may not be worth the time and opportunity cost.
  - wsinks 7y ago
    Smart writeup! I'm definitely following some people that I know are 'pitching their dream' (like Peter Zeihan) but have access to useful data and make opinions. I definitely think that specific example is in the hot take category. Do you have any people you'd like to call out to follow? I'm curious about good environmental data and how we can better model our investment in our natural resources.
    - SkyMarshal 7y ago
      Afraid I don't have any recommendations for your interests, but this kind of network is relatively personal and best cultivated on your own, without relying on others' recommendations, imho.
jeffadotio 7y ago
One becomes so caught up in the fear of becoming obsolete and the dopamine-fueled satisfaction of staring into a screen that we fail to make simple evaluations that are right in front of us. Is a thumbs up emoji posted by someone with whom you shook hands at a conference six months ago while he sits on the toilet going to give you any real value or are you just succumbing to social pressure? In the thousands of hours that users sink into social media in a year one could instead read, learn to code, learn an instrument or learn anything at all. Who is more interesting, a biologist who plays the mandolin and is semi-fluent in a second language or a biologist with lots of Twitter followers?
tsukurimashou 7y ago
J. Peterson suspects the small character limit get people to react impulsively and often aggressively.
- mc32 7y ago
  Imagine the instant plunge in usership if they required a (word/char) minimum and actively screened for filler/mumbojbo/word inflation with whatever ML tools they have.
- aaron695 7y ago
  Twitter is a cesspool of shit. The worst of humanity. But like any war zone if you're tough and willing to take a risk there's money to be made. But if you cry over getting rape and death threats or don't like people who make them, then just don't go there. That is part of Twitter, it's in it's fundamental design, you enter 7 billions peoples homes and ask what they think of your opinion. It's not for the normal scientist, they can do great things outside of Twitter. To your point, yes the character limit seems likely it is a factor of why it's so bad. But lot's more people than J. Peterson think that. But he perhaps was one of the first to think fundamentally at a low physiological level what was happening.
buboard 7y ago
Please dont! Very very few scientists can use twitter for scientific discussions, they are interesting, technically capable, and are already there. Do not encourage scientists to slide down the road of self-promotion, as most people on twitter do. I hate how much time academic institutions and PIs are investing in social media, curating facebook pages, twitter pages, a linked pages yadda yadda. It's mostly scientists chasing publicity so they can add another blog/newspaper article bullet to their CV so they can get more funding. I sincerely hope funding bodies start banning those publicity bullet points from CVs during evaluations.
- Fomite 7y ago
  My field's Twitter hashtag is one of the best sources for interesting and technically capable discussions. If I have a question, at this point I'll turn to Twitter before I turn to say...the relevant Stack Exchange site. Also, most major funding bodies (NSF, NIH, etc.) use standard "biosketch" templates that are separate from CVs already and don't use that category.
xhkkffbf 7y ago
When do we get to mob people we don't like? It's like covertly campaigning against tenure but without being passive aggressive! I'm just here for the ratio!
andreyk 7y ago
I am a PhD student at Stanford (focusing on AI/learning algorithms for robotics manipulation) and happened to retweet this article, so perhaps I can give my perspective. This is what I tweeted about this: "Neat discussion of benefits of this here platform for scientists / researchers. Personally I was pretty mystified with its popularity til trying it and finding it useful, so this might be helpful for like-minded ppl. But, imo it's really mostly a great way to keep up w news. https://t.co/J3aLAh56W3" https://t.co/J3aLAh56W3" Really, don't knock it til you try it. As I said in a reply on here, the biggest benefit is being able to keep up with neat new papers from researchers you think do cool work. Just after that first tweet I just happened to see this from a researcher at Google: "New from Google Research! REALM: https://t.co/kS2oTyxAAj https://t.co/kS2oTyxAAj We pretrain an LM that sparsely attends over all of Wikipedia as extra context. We backprop through a latent retrieval step on 13M docs. Yields new SOTA results for open domain QA, breaking 40 on NaturalQuestions-Open! https://t.co/DYDFX69Td8" https://t.co/DYDFX69Td8" And I retweeted it with a comment that it's cool (and read the paper to understand a bit more) and now follow that researcher. Like any big platform, Twitter can be used in bad ways (engaging in drama, constantly looking at it instead of working coughmuch like HNcough, etc.), and it can be used in many good ways. I was pleasantly surprised with how useful it actually is, personally. That being said... much of the content of this article is kinda weird, it's not like this is that complicated... I guess the intent is to speak to more senior researchers, but it does come off weird.
- jeffadotio 7y ago
  > it's really mostly a great way to keep up w news Receiving information is not the same as being informed. The head of Facebook’s news authority is the owner of a far-right news site. A famous 2012 study[1] found that Fox News viewers were the least informed on facts covered by the news while NPR listeners were the most informed. Fox viewers were even less informed than those who consumed no news media at all. All news is not created equal and you are what you read. Curation is important and an organization with a fact-oriented status quo is a superior source. 1. https://www.poynter.org/reporting-editing/2012/survey-nprs-listeners-best-informed-fox-news-viewers-worst-informed/ https://www.poynter.org/reporting-editing/2012/survey-nprs-l...
dekhn 7y ago
Scientists should spend less time on twitter. It's all noise, and even the signal gets distorted by all the crazies.
codekilla 7y ago
Don’t
senderista 7y ago
There's definitely a lot of good content one can find linked on Twitter (not so much tweets themselves). But I ultimately decided it wasn't worth the opportunity cost when I have a house full of books I could be reading instead.
hirenj 7y ago
Within my field, it looks like science twitter is useful, but only up to a limit. The primary use derived from it is a list I found myself added to. It actually contained people who were only peripherally related to the research I do, but meant that there was great diversity in the papers that were promoted in the feed - not all relevant, but at least interesting. A second use is a signalling factor that you are active in the field - really an exercise in building brand. Finally, there are probably two people who actively engage in discussion on Twitter (from my field). This doesn't happen more often because scientific review is a subtle beast, and doing it in the open with 280 chars, as quickly as possible on a phone isn't conducive to constructive feedback. In fact, I've got a number of draft tweets initiating discussions that I didn't end up sending because it just felt plain rude to review in that manner!

96 points by amrrs 7y ago

One thing that this article misses about twitter is that you carry your followers around with you. So if you have 1000 followers and you tweet at someone, whatever you're saying, you're saying in front of 1000 of your fans. So if you disagree with someone not only are they going to get your criticism, they're likely to get 1000 people piling on behind you. There are lots of things I can say to you one on one that are suddenly basically bullying if I've got a mob behind me. Equally, if you draw attention to someone, even if you aren't disagreeing with them, you have to be aware that you're likely subjecting them to a level of unwanted attention from your audience.

bosswipe 7y ago
If you reply to someone's tweet then mostly the people who follow both accounts will see the reply, in addition Twitter's algorithms will opaquely decide whether to show the reply to your other followers. If you want to ensure that your followers see the reply you can use Retweet with Comment.

azangru 7y ago

> There is no single guideline on when to block somebody—some may only support blocking trolls, and refrain from blocking other scientists who may be critical, while others believe it’s perfectly fine to mute or block any account... A good rule of thumb might be to block somebody if they have a repeated negative effect on your Twitter experience > It is perfectly okay to not look at Twitter for hours or even days. > Do not worry that this might be confusing: Twitter users can easily skip over tweets in languages they are not proficient in... Who is the target audience of a text like this? Do people feel they need someone's permission to block someone, or to skip Twitter for a while, or to write in whichever language they please? Honestly!

thrwaway69 7y ago
The point is to commoditize yourself so that's not unexpected. The brand (you) has to get good at targeting the audience they are selling product (yourself) to and avoid interacting with the competition or getting run over by trolls. Protecting the brand (yourself) is important. To distinguish the brand (yourself) from others, the brand (you) has to set something people can look up to as a way to differentiate their experience with them (you). What the brand (you) likes, follows, engages with etc defines the image of them (you) and the right market fit (their feed). Accessibility is a huge factor for a commodity (sellable parts of you as a whole) whilst some blocking may be necessary for setting the cultural expectations for the brand (you). If the brand (you) block people from certain group and it is public, that might work as a signal for what the brand (you) stands for. Let's suppose the brand (I) blocks a mark Zuckerberg and Facebook, then that works as a signal for that the brand (I) cares about privacy and against people who exploit that. If the brand (I) only uses English to tweet, then that might be a signal for which audience product (I) is made for. If the brands (yours) block list isn't diverse, it might signal as a racial/ethnic/gender/sex bias.
im3w1l 7y ago
It's not about "permission". It's about popularity. If you write too many posts in languages your followers don't speak, they might curate you out of their feeds decreasing your popularity. As a newcomer you don't know how important it is to take this into account.
Mountain_Skies 7y ago
The president was sued for blocking people on Twitter and ordered by a court to not block anymore. That might create some confusion especially for anyone making a Twitter account as an official communication medium for a government facility.

knzhou 7y ago

As a scientist I have a strong aversion to getting on Twitter, and this article isn't helping. All my experience with Twitter has told me it's a concentrator for hot (i.e. wrong) takes, snarky dismissals, and sloganeering, to a greater extent than any other social media platform. It seems to degrade thoughtful people into ordinary ones and ordinary people into monsters. I'll take even HN flamewars over that any day.

SkyMarshal 7y ago
>It seems to degrade thoughtful people into ordinary ones and ordinary people into monsters. In my experience that’s about 90% true, but there’s a small minority of people who remain thoughtful and insightful even in 280 characters. If you can find them and mutually follow each other, and then mute or block all the blowhards, tryhards and self-promoters, you’ll have cultivated a useful network of colleagues beyond what you might otherwise have via work and conferences. And if multi-disciplinary perspectives are valuable to your own research, then that Twitter network is especially valuable in that regard, as it’s likely to be at least as multidisplinary as your day-to-day work environment and more-so than your usual conference circuit. That said, it does take some work to cultivate that graph. May or may not be worth the time and opportunity cost.
- wsinks 7y ago
  Smart writeup! I'm definitely following some people that I know are 'pitching their dream' (like Peter Zeihan) but have access to useful data and make opinions. I definitely think that specific example is in the hot take category. Do you have any people you'd like to call out to follow? I'm curious about good environmental data and how we can better model our investment in our natural resources.
  - SkyMarshal 7y ago
    Afraid I don't have any recommendations for your interests, but this kind of network is relatively personal and best cultivated on your own, without relying on others' recommendations, imho.

jeffadotio 7y ago

One becomes so caught up in the fear of becoming obsolete and the dopamine-fueled satisfaction of staring into a screen that we fail to make simple evaluations that are right in front of us. Is a thumbs up emoji posted by someone with whom you shook hands at a conference six months ago while he sits on the toilet going to give you any real value or are you just succumbing to social pressure? In the thousands of hours that users sink into social media in a year one could instead read, learn to code, learn an instrument or learn anything at all. Who is more interesting, a biologist who plays the mandolin and is semi-fluent in a second language or a biologist with lots of Twitter followers?

tsukurimashou 7y ago

J. Peterson suspects the small character limit get people to react impulsively and often aggressively.

mc32 7y ago
Imagine the instant plunge in usership if they required a (word/char) minimum and actively screened for filler/mumbojbo/word inflation with whatever ML tools they have.
aaron695 7y ago
Twitter is a cesspool of shit. The worst of humanity. But like any war zone if you're tough and willing to take a risk there's money to be made. But if you cry over getting rape and death threats or don't like people who make them, then just don't go there. That is part of Twitter, it's in it's fundamental design, you enter 7 billions peoples homes and ask what they think of your opinion. It's not for the normal scientist, they can do great things outside of Twitter. To your point, yes the character limit seems likely it is a factor of why it's so bad. But lot's more people than J. Peterson think that. But he perhaps was one of the first to think fundamentally at a low physiological level what was happening.

buboard 7y ago

Please dont! Very very few scientists can use twitter for scientific discussions, they are interesting, technically capable, and are already there. Do not encourage scientists to slide down the road of self-promotion, as most people on twitter do. I hate how much time academic institutions and PIs are investing in social media, curating facebook pages, twitter pages, a linked pages yadda yadda. It's mostly scientists chasing publicity so they can add another blog/newspaper article bullet to their CV so they can get more funding. I sincerely hope funding bodies start banning those publicity bullet points from CVs during evaluations.

Fomite 7y ago
My field's Twitter hashtag is one of the best sources for interesting and technically capable discussions. If I have a question, at this point I'll turn to Twitter before I turn to say...the relevant Stack Exchange site. Also, most major funding bodies (NSF, NIH, etc.) use standard "biosketch" templates that are separate from CVs already and don't use that category.

xhkkffbf 7y ago

When do we get to mob people we don't like? It's like covertly campaigning against tenure but without being passive aggressive! I'm just here for the ratio!

andreyk 7y ago

I am a PhD student at Stanford (focusing on AI/learning algorithms for robotics manipulation) and happened to retweet this article, so perhaps I can give my perspective. This is what I tweeted about this: "Neat discussion of benefits of this here platform for scientists / researchers. Personally I was pretty mystified with its popularity til trying it and finding it useful, so this might be helpful for like-minded ppl. But, imo it's really mostly a great way to keep up w news. https://t.co/J3aLAh56W3" https://t.co/J3aLAh56W3" Really, don't knock it til you try it. As I said in a reply on here, the biggest benefit is being able to keep up with neat new papers from researchers you think do cool work. Just after that first tweet I just happened to see this from a researcher at Google: "New from Google Research! REALM: https://t.co/kS2oTyxAAj https://t.co/kS2oTyxAAj We pretrain an LM that sparsely attends over all of Wikipedia as extra context. We backprop through a latent retrieval step on 13M docs. Yields new SOTA results for open domain QA, breaking 40 on NaturalQuestions-Open! https://t.co/DYDFX69Td8" https://t.co/DYDFX69Td8" And I retweeted it with a comment that it's cool (and read the paper to understand a bit more) and now follow that researcher. Like any big platform, Twitter can be used in bad ways (engaging in drama, constantly looking at it instead of working coughmuch like HNcough, etc.), and it can be used in many good ways. I was pleasantly surprised with how useful it actually is, personally. That being said... much of the content of this article is kinda weird, it's not like this is that complicated... I guess the intent is to speak to more senior researchers, but it does come off weird.

jeffadotio 7y ago
> it's really mostly a great way to keep up w news Receiving information is not the same as being informed. The head of Facebook’s news authority is the owner of a far-right news site. A famous 2012 study[1] found that Fox News viewers were the least informed on facts covered by the news while NPR listeners were the most informed. Fox viewers were even less informed than those who consumed no news media at all. All news is not created equal and you are what you read. Curation is important and an organization with a fact-oriented status quo is a superior source. 1. https://www.poynter.org/reporting-editing/2012/survey-nprs-listeners-best-informed-fox-news-viewers-worst-informed/ https://www.poynter.org/reporting-editing/2012/survey-nprs-l...

dekhn 7y ago

Scientists should spend less time on twitter. It's all noise, and even the signal gets distorted by all the crazies.

codekilla 7y ago

Don’t

senderista 7y ago

There's definitely a lot of good content one can find linked on Twitter (not so much tweets themselves). But I ultimately decided it wasn't worth the opportunity cost when I have a house full of books I could be reading instead.

hirenj 7y ago

Within my field, it looks like science twitter is useful, but only up to a limit. The primary use derived from it is a list I found myself added to. It actually contained people who were only peripherally related to the research I do, but meant that there was great diversity in the papers that were promoted in the feed - not all relevant, but at least interesting. A second use is a signalling factor that you are active in the field - really an exercise in building brand. Finally, there are probably two people who actively engage in discussion on Twitter (from my field). This doesn't happen more often because scientific review is a subtle beast, and doing it in the open with 280 chars, as quickly as possible on a phone isn't conducive to constructive feedback. In fact, I've got a number of draft tweets initiating discussions that I didn't end up sending because it just felt plain rude to review in that manner!