8:00 AM–9:00 AM
Wake up, eat breakfast, pack lunch, leave the apartment. On the bus, open the RSS reader on the phone and skim the “life” folder for news or blog posts worth noticing. Most of it gets glanced at and forgotten; anything long goes straight into Pocket for the evening.
The folder for paper alerts is a different matter. That one usually doesn’t get fully dealt with until after the Friday group meeting, even though the unread count has already passed 100+. In practice, journals like Science, Nature, and PNAS rarely publish papers directly in my niche, and if they do, I’ll see them through other channels anyway. RSS is mainly there to track the top three journals in my field plus keywords tied to my own research.
Roughly speaking: I look at the title and graphical abstract for essentially 100% of the papers that come in, read the abstract for maybe 20–50%, inspect the figures for 5–10%, and read the full paper for 1–3%.
9:00 AM–9:30 AM
Get to the lab, say hi to the postdocs, PhD students, and master’s students, then open Slack. The professor assigned another task overnight and shared five newly published papers: three tagged ideas, two tagged experimental design. Today also includes a progress discussion with a PhD student on Project A, plus internal training for a new master’s student.
At least the training materials already exist in the lab’s shared Dropbox folder, so they can be sent over immediately. Not that she’ll actually read them.
Then it’s time to check my own channel: one peer review needs to be finished today, the next manuscript draft needs revision based on the last one-on-one meeting with the professor—which happens every two weeks—and this afternoon I need to prepare standards for an experiment scheduled two days from now.
9:30 AM–11:00 AM
Peer review.
To be fair, the paper is genuinely interesting. The main problem is the data processing: something about it feels off, the fitted curves are unexplained, and it’s hard to tell what model was even used. The electron microscopy images are too blurry, and some extra experiments would help.
The authors did at least share their raw data on figshare. I ran a script on it and confirmed that the results match what appears in the paper, which is a real point in their favor.
On average, a review takes about six hours: 1.5 hours to understand the paper, 3 hours to check the details, and 1.5 hours to write the report. In reality, that usually gets spread across five days. All comments go into a matching Google Doc so nothing gets lost.
As a rule, either review quickly or don’t agree to review at all. Other people are waiting on you.
And if you do review, keep the record. Publons matters because review history is part of academic reputation now. Post-publication commenting platforms are also worth watching; open commentary after publication is likely to become a bigger part of how researchers exchange ideas.
11:00 AM–12:00 PM
Internal training for the new master’s student.
The session covers how the lab uses Slack and where resources live, the standardized presentation style and lab logo for external talks, preferred data-analysis models and why the group favors them, reimbursement and finance procedures, and so on.
Because training materials have been built up over time, the expectation is simple: learn the lab’s internal language within a week, then give a short background presentation at the next group meeting as a check.
Lab turnover is high enough that repeating the same training from scratch can quietly consume absurd amounts of time. Better to archive old workshop material and previous training sessions as self-study resources, then build in some kind of checkpoint. Everyone is an adult. No one needs to be begged into learning.
Training newcomers matters, but universities and institutes already send around plenty of general training announcements. Forward those in real time. If someone intends to do more than coast through a degree, they should know what to do with them.
New researchers also need to take ownership of their online academic presence. Keeping profiles updated on platforms like ResearchGate or Academia, participating in academic Q&A communities, and maintaining profiles on Google Scholar or Baidu Scholar all contribute to visibility. Search priority alone makes those profiles worth maintaining.
12:00 PM–1:30 PM
Lunch.
After eating, check email. Someone is asking about details of data processing from an earlier paper, so I find the old project’s GitHub repository and send it over. Those materials were always going to be public after publication anyway.
A few groups I follow on ResearchGate have also published new papers. Some overlap with what the professor already sent in the morning, and some are updates from projects I’ve been tracking. Meanwhile, a few big names I follow on bioRxiv have posted new preprints.
I skim, make notes, and retweet anything especially interesting. This still counts as a break. No deep reading required; the point is just to register what’s happening.
When replying to email, brevity is usually best. There’s no need to drown everything in politeness. Email is already being displaced, little by little, by faster communication channels. Focus on the actual issue. Any task mentioned in email gets turned immediately into a calendar item with a reminder, and then it leaves my head.
1:30 PM–2:30 PM
Project meeting with a PhD student.
The results are promising overall, though a few additional experiments are needed. The conversation is not face-to-face because the student is currently visiting another university. Everything happens through video chat and a shared collaborative document. We talk while editing the same report, using voice input where convenient, and by the end of the discussion the report is basically drafted.
Then it gets shared with the professor, and the update is logged in the project channel on Slack.
2:30 PM–3:30 PM
Manuscript revision.
Writing happens in RStudio with R Markdown. Using templates from the rticles package, it’s easy to generate both TeX and PDF output directly. Reference management goes through Zotero because collecting papers online is painless that way.
If someone prefers a what-you-see-is-what-you-get workflow, Google Docs with Paperpile works too. Word plus EndNote is still perfectly usable if that’s your habit. The exact tools matter less than having a system.
The literature library should be updated weekly and organized by project. That alone saves time.
Once a manuscript is finished, posting it to a preprint server can be a sensible move. It won’t solve everything, but it can reduce the risk of being scooped while the paper is stuck in review.
3:30 PM–5:30 PM
Prepare for the experiment scheduled in two days.
This is the physically tedious part, and usually the biggest time sink of the day, mostly because something unexpected always happens.
Don’t assign an entire day to one thing if you can help it. Assign one hour to one thing. Focus hard, then move.
5:30 PM–6:30 PM
Record what got done today and what needs doing tomorrow in Slack. Set up discussion times with other people. Check email for the second time and send replies.
6:30 PM–8:00 PM
Cook, eat, and listen to the articles saved in Pocket using read-aloud mode.
8:00 PM–10:00 PM
This block goes to online courses, reading, checking whether academic Twitter has erupted over something, or updating a personal blog with a note about a newly published paper.
It doesn’t need to be the same every night. Studying, wandering, tinkering—any of that is fine. If you keep a long-term habit of learning and exploring, you eventually notice that truly new knowledge does not arrive in huge daily waves. Most of what circulates is repetition in slightly different packaging.
Researchers need to know how to work with the internet, not just around it. Having a personal website or a lab website helps, as does keeping an online CV. But online presence doesn’t have to be about chasing prestige. It can also be about sharing knowledge well: writing systematic overviews of frontier problems, turning them into continuously updated books, publishing web-based teaching slides, or posting conference presentations where others can actually find them.
A researcher is still just a person first. When something comes up, searching Google or Baidu is normal. If people keep running into your work online, that visibility becomes part of your reputation whether you intended it or not.
And while most principal investigators are still over 40 now, eventually leadership will belong to generations that grew up on the internet. Their habits of learning and evaluation already depend heavily on online resources. Communities like Quora, Stack Overflow, Reddit, Zhihu, Guokr, and ScienceNet are not side shows; they quietly accumulate professional reputation.
It is entirely possible for someone to become known online long before joining an elite institution. I’ve seen a lab at MIT recruit a postdoc and publicly welcome them as a guru in their area—and that person had already built a formidable reputation in online communities during their PhD, even though their doctoral institution was nowhere near MIT’s level.
10:00 PM–12:00 AM
Actual downtime.
Play a game, work out, watch a movie, scroll social media, like a few posts from friends. This is the buffer zone before sleep.
The internet has been quietly reshaping one industry after another, and research is no exception. Using the right tools well can improve efficiency and reduce anxiety.
Sometimes I think forward-looking labs should probably run their own public channels more seriously: not only to share their results, but also to highlight important new papers in the field. For many people, following a steady stream of curated updates is easier than sitting down with a review article. Over time that kind of habit can build a broader view of a discipline and even influence what people regard as the frontier. In some ways, that can be more effective than waiting for the once-a-year or once-every-two-years conference to hear a few talks.
Outside China, academic Twitter and hashtags already do much of this work. There are even sites that try to quantify researchers’ social-media influence. That kind of influence should not be dismissed. The people who fund research may read journals, but they definitely also scroll their feeds.
Anyway, take all of this with an appropriate amount of skepticism. If any of it sounds familiar, then it sounds familiar.