this post was submitted on 25 Aug 2024

869 points (98.9% liked)

Science Memes

10340 readers

1888 users here now

Welcome to c/science_memes @ Mander.xyz!

A place for majestic STEMLORD peacocking, as well as memes about the realities of working in a lab.

Rules

Don't throw mud. Behave like an intellectual and remember the human.
Keep it rooted (on topic).
No spam.
Infographics welcome, get schooled.

Research Committee

!spiders@lemmy.world

Other Mander Communities

Science and Research

Biology and Life Sciences

Physical Sciences

Humanities and Social Sciences

Practical and Applied Sciences

Memes

Miscellaneous

founded 2 years ago

MODERATORS

Sal@mander.xyz

fossilesque@mander.xyz

SciBot@mander.xyz

869

Use Zotero (mander.xyz)

submitted 3 weeks ago by fossilesque@mander.xyz to c/science_memes@mander.xyz

49 comments fedilink hide all child comments

all 50 comments

sorted by: hot top controversial new old

[–] Got_Bent@lemmy.world 52 points 3 weeks ago* (last edited 3 weeks ago) (1 children)

I'm so old that the last time I wrote a research paper, it was on a word processor with no Internet connection or spell check.

Given such constraints, I can't fathom the concept of waiting until the end to add all the references. If I didn't do it as I went, I'd have surely died.

I should add that we always read each other's papers before submission to get a second set of eyes for errors, misspellings, and grammatical quagmires. It was mutually beneficial as the reviewing made us all better. Is that still a common practice?

[–] phdepressed@sh.itjust.works 17 points 3 weeks ago (1 children)

Yes having friends/colleagues read papers before submission is still common practice.

[–] Got_Bent@lemmy.world 4 points 3 weeks ago (2 children)

I hope that wasn't a stupid question. I've been out of that game since before most people on the Internet were born, so I didn't want to make assumptions based on "that's how we used to do it."

[–] phdepressed@sh.itjust.works 4 points 3 weeks ago

I think that type of feedback is a forever thing. How it may be done can change (even some PIs just do comments in the file instead of printing and handwriting) but I don't see a scenario when that isn't useful anymore.

[–] blanketswithsmallpox@lemmy.world 0 points 3 weeks ago (1 children)

Something something chatgpt something something boomer skibidi.

[–] three@lemm.ee 3 points 3 weeks ago

sigma ohio vibes desu

[–] Sterile_Technique@lemmy.world 38 points 3 weeks ago (4 children)

Use Zotero

...jesus-tapdancing-christ-on-a-cracker, batman! How have I never heard of this?! I just pulled up a few videos on Zotero and this shit looks amazing!

I'm a semester into nursing school, and I know the higher we push that degree up, the more writing - and more strict with citations - the content becomes.

I'm downloading the shit out of this.

Thank you so much for making that your title!!

[–] brbposting@sh.itjust.works 15 points 3 weeks ago

Open source and looks very useful so I’ll link

[–] shikitohno@lemm.ee 15 points 3 weeks ago (1 children)

Zotero is awesome, and I'd also recommend the browser extension with an account. Account is free, but it'll let you save any web sources with all the metadata you need and sync it to the main program.

Zotero in combination with LaTeX and Biber have saved me so much time. Especially when I had a crazy professor who would make last minute changes to style requirements. I remember we had a paper to write that they initially said "Just cite with whatever format you want, it's fine as long as they info is there." Cool, write my ten pages or whatever with footnotes containing short citations and full citations available in the bibliography at the end. The night before the paper was due, "Actually, I need all papers to use APA citations or you'll be docked points.". No problem, just change my citation style at the top of my LaTeX doc and tell it to reprocess the paper, all the citations fixed in about 5 seconds, without even needing to learn the ins and outs of APA formatting.

[–] Sterile_Technique@lemmy.world 7 points 3 weeks ago

Dude it is blowing my mind how good this software is. This semester has like 30 or so different drugs we need to read up scattered in these weekly worksheets. All the info we plug in needs to be cited. Was the same last semester and those friggin things took FOREVER to finish. This semester... I've been doing pretty much nothing but those since my last post. They're done. For the entire semester... which just started. They're fucking DONE! All the in-text citations... *poof* there they are! All the full citations at the end of the document... *poof* all there, correctly formatted and in the correct order.

That was days of work last semester. I just got it all done in 6 hours.

BRUH!!

[–] Omniraptor@lemm.ee 9 points 3 weeks ago* (last edited 3 weeks ago)

It's kinda fucked up that they don't teach this to freshman college students first thing as part of basic media literacy.

Same goes for teaching version control such as git to anyone writing or collaborating on complex text (not necessarily even code). https://medium.com/@RichHosek/github-for-writers-an-introduction-8cec9d9ece2

Granted, Google docs already has built in versioning, but git works better offline and works with any text editor.

[–] bitfucker@programming.dev 7 points 3 weeks ago (1 children)

Use zotfile extension (or equivalent since the v7 is not backward compatible) and it'll save your time managing those pesky pdf too.

[–] Sterile_Technique@lemmy.world 3 points 3 weeks ago

Thanks for the tip - will do!

[–] Track_Shovel 24 points 3 weeks ago

I write monstrous documents (hundreds of pages). Zotero is all I use. It's ridiculously simple and you CAN SHARE REFERENCE WITH OTHER PEOPLE SO THEY DONT HAVE TO ENTER THEM.

This might not seem like a big deal, but I can get junior to prep a doc and share reference, or just enter references for me before I start writing and it saves a pile of time. Further, I can call references that my colleagues have entered.

[–] Engywuck@lemm.ee 18 points 3 weeks ago (1 children)

My boss usually adds references by hand. I think he's just crazy.

[–] watersnipje@lemmy.blahaj.zone 6 points 3 weeks ago (1 children)

Jesus. Does he work by candlelight too?

[–] Engywuck@lemm.ee 2 points 3 weeks ago

Yeah, it' pretty surprising to me that people I consider so smart are technologically inept.

[–] kireotick@lemmy.world 17 points 3 weeks ago

Yeah as an assistant for a bunch of 1st yearers at university I am sooo going to recommend Zotero

[–] Imgonnatrythis@sh.itjust.works 13 points 3 weeks ago (2 children)

It's pretty much abandonware, but the original mendeley software is the bomb. Starting to play with afforai and it seems interesting.

[–] bitfucker@programming.dev 6 points 3 weeks ago (1 children)

Uhhh, they just released the v7 tho?

[–] Imgonnatrythis@sh.itjust.works 6 points 3 weeks ago

The new one is severely crippled. It's just a web shell. I forget when they switched over but I think it was when Elselvier bought them. At the same time they dropped their phone app which was a real bummer. It's obvious that it's more of a side project now than a serious app.

[–] scrooge@infosec.pub 2 points 3 weeks ago

Yup, messed up all my workflow when they changed the plugin

[–] vzq@lemmy.blahaj.zone 8 points 3 weeks ago (1 children)

I mean, depends on whether you think someone will actually check what the references actually say or just eyeball the titles and go LGTM.

[–] 474D@lemmy.world 3 points 3 weeks ago

Lol this. Back in college, I would just write the paper and then grab some sources from the bottom of some corresponding Wikipedia pages

[–] Greyghoster@aussie.zone 5 points 3 weeks ago (1 children)

That’s the same as writing tests for software code after writing the code. A mystery without end or reliability.

[–] smpl@discuss.tchncs.de 9 points 3 weeks ago (3 children)

Tests? What tests?

[–] MonkderVierte@lemmy.ml 6 points 3 weeks ago (1 children)

printf

[–] smpl@discuss.tchncs.de 4 points 3 weeks ago

I have this in code I'm writing right now...

#ifdef DEBUG
    #define DEBUG_PRINT(...) printf(__VA_ARGS__)
#else
    #define DEBUG_PRINT(...)
#endif

It is the most straighforward way to get the state of things while hammering on the keyboard trying to mash up something that looks like a program.

[–] null 5 points 3 weeks ago

The software either works or it doesn't. There, tested!

[–] Greyghoster@aussie.zone 2 points 3 weeks ago

True, they can be hard to find.

[–] darki@lemmy.world 5 points 3 weeks ago* (last edited 3 weeks ago) (1 children)

Mendeley also works ok 🫡 but I still add them mostly in the end lol just need yo put " [that really important paper]" and then do it after

[–] Omniraptor@lemm.ee 5 points 3 weeks ago* (last edited 3 weeks ago)

If anyone is choosing between them I'd advise going with Zotero because it's an open source project. No chance of it suddenly shutting down or changing the terms of service.

[–] HubertManne@moist.catsweat.com 0 points 3 weeks ago (2 children)

thing about that is references are added and taken away as you change the paper, read something else, whatnot. What I start with and what I end with can be pretty much separate papers.

[–] fossilesque@mander.xyz 4 points 3 weeks ago

Use latex.

[–] nossaquesapao@lemmy.eco.br 4 points 3 weeks ago (1 children)

But zotero deals with that just fine. If you remove a citation from the text, the reference at the end will be removed as well. and each new citation will be added normally.

[–] HubertManne@moist.catsweat.com 2 points 3 weeks ago

Oh yeah sorry. I did not mean to say the software can't handle it. Its more psychological for me like I want to do it when the paper is where it should be and lets forget about those roads we went down we shall not speak of :)

[+] sturlabragason@lemmy.world -17 points 3 weeks ago* (last edited 3 weeks ago) (3 children)

This is what LLMs are for. Just double check their work.

Edit: It seems based on the many downotes that I don’t know shit about citations 😅

Guys, I have no idea, never had to write one of these. 😅 It was just a suggestion.

I use GPT4o daily for work stuff. It often does the job pretty reliably. However all the inserting citation into the text would of course not work, seems that zotero that others are mentioning is made for the job and would therefore do it much better.

It seems people let their prejudices get in the way of using LLMs like the limited tools they are. They are not AI, just a pretty decent guessing machine with multiple limitations.

[–] fossphi@lemm.ee 14 points 3 weeks ago (1 children)

How exactly would you use an LLM for something like this?

[–] sturlabragason@lemmy.world -2 points 3 weeks ago (5 children)

Here’s an example prompt;

Prompt for LLM on Generating APA Citations (7th Edition)

Instructions: I need to create APA format citations for my research paper based on the 7th edition of the APA guidelines. Below are examples of correct APA citations for different types of sources, followed by a list of sources that need to be cited. Please generate the citations accurately, paying special attention to formatting rules such as capitalization, italicization, and punctuation.

Important Formatting Rules:

Titles: For journal articles and web pages, use sentence case (capitalize only the first word of the title and subtitle, if any, and any proper nouns). For books and journals, use title case (capitalize all major words) and italicize the title.
Authors: List up to 20 authors in the citation. For more than 21 authors, list the first 19 followed by an ellipsis and the final author's name.
DOIs and URLs: Always include the full DOI or URL as clickable links without a period at the end.

Citation Examples:

Journal article: Author, A. A., Author, B. B., & Author, C. C. (Year). Title of the article. Title of Periodical, volume number(issue number), pages. https://doi.org/xx.xxx/yyyy
Book: Author, A. A. (Year). Title of the book. Publisher.
Website: Author, A. A. (Year, Month Day). Title of the webpage. Website Name. URL

Sources List:

Smith, J. (2022). Understanding LLMs. Journal of AI Research, 15(2), 123-145. https://doi.org/10.1016/j.jair.2022.05.001
Brown, L., & Green, T. (2019). AI in modern education. Cambridge University Press.
Miller, S. (2021, June 5). The evolution of language models. Tech Times. https://www.techtimes.com/evolution-language-models
Johnson, P., Lee, M., & Black, R. (2020). The future of AI in society. AI and Ethics Review, 22(4), 215-230.

Additional Instructions:

Ensure that article titles are in sentence case, while journal names and book titles are in title case and italicized.
For multiple authors, follow APA guidelines and use "&" between authors, with commas separating names.
Make sure DOIs and URLs are accurate and correctly formatted as clickable links.

Once you have generated the citations, please double-check for consistency in formatting (capitalization, italics, and punctuation) and ensure that all sources follow the APA 7th edition rules.

Thank you!

[–] nossaquesapao@lemmy.eco.br 10 points 3 weeks ago

But why do it like that, if zotero is easier to use, less prone to errors and uses much fewer resources?

[–] phdepressed@sh.itjust.works 10 points 3 weeks ago (1 children)

This only builds a formated bibliography where your input is an unformated bibliography... It also can't accurately do the more tedious (and important)part of actually putting the in text citations. In short you have to do the same amount of upfront work of getting all the relevant citations in a list then additional work on checking that the llm formatted properly and didn't hallucinate then still have to go put in the in text citations.

Wheres with endnote or zotero you choose from your list add the intext, autoformat from your choice and it's automatically put in a formatted bibliography.

[–] sturlabragason@lemmy.world 2 points 3 weeks ago

Yeah from my perspective I somehow assumed that was the hard part 😅

[–] Vlyn@lemmy.zip 10 points 3 weeks ago (2 children)

Good joke, you'd have to double and triple check every single citation as LLMs love to hallucinate.

I wouldn't risk my title or expulsion because an LLM fucked up my work. Even one missed reference or a false citation could cost you plenty.

[–] sturlabragason@lemmy.world 1 points 3 weeks ago

Yeah newer models are less prone to that problem but definetely agree 🙂

[–] sturlabragason@lemmy.world 1 points 3 weeks ago

The intentiom was for the author to supply the data and the LLM to provide the formatting only.

[–] fossilesque@mander.xyz 2 points 3 weeks ago

This will not get you far lol.

[–] smpl@discuss.tchncs.de 8 points 3 weeks ago

That's a lot of trouble, you can just ask it if it's telling the truth.

[–] Atrichum@lemmy.world 3 points 3 weeks ago (1 children)

It straight up makes up sources and citations.

[–] sturlabragason@lemmy.world 2 points 3 weeks ago

I was intending for the author to chomp in the data and the LLM to spit it out properly formatted.