Public developer API

Feature requests
,
21

Yes, an API enabling things like IFTTT integration would be great.

22

Is there any update on this? It’s been five years since this feature was requested, and in my mind it’s the missing feature that’s preventing Paperpile from being a truly great citation manager. It gets so tiring when writing a LaTeX article to click “export to bibtex” each time I add a new citation to a paper, download the file, and then move it to the proper directory and rename it.

3 Likes
23

We are painfully aware that BibTeX workflows are not perfect and there is much room for improvement. But we see this independent of the topic of this post, i.e. a public developer API. We invested quite a bit in our Word plugin over the past years. Although it’s now released as “Word plugin” it is essentially a Desktop client that can sync your complete library to your desktop in the background.

From there, our goal is to transparently write BibTeX files that are always up-to-date and also allow you to create and insert Latex citations in a truly “cite while you write” manner.

1 Like
24

I’m a scientist and bioinformatician, and doing this manual is losing time. I’m using Pocket to mark papers as to_read, which uses IFTTT to add a task in Todoist. Then I have to manually add it to Paperpile and star it, which will lead it to being synced to my Remarkable e-reader. I hate manual parts and will automate anything, but Paperpile is still lacking any options for this it seems. I love paperpile and have been using (and converting others) for years, but at this point I may have to start searching for an alternative to fix this missing link in my workflow.

25

There are a lot of people with idle time this summer, can we make this a sprint of some kind? I love paperpile of but the lock in is making somethings stale and brittle. Would love to be able to search my library without exporting. It’s already a web interface we should be able to do this.

2 Likes
26

1+ - How about this idea - Make PP as much of an open source platform as possible. For example, make the PP database format available. That might allow outsiders to create an API, among other things.

3 Likes
27

Hi Stefan,

I wonder if there is an update on an API to fetch all of paperpile records as bibtex? I am using Zettlr note taking tool on my macbook but prefer to use years of collection in paperpile for citations.

Thanks,
Samir

1 Like
28

Yes, we should have something to share relatively soon related to automatic BibTeX export via an API endpoint.

In case anyone is interested here, we also have put together a short survey to better understand the LaTeX and BibTeX use cases of our users:

30

Just want to second the request for a public api. Looked at your survey and what you all have planned. Looks really nice, but it does limit the flexibility. An api allows workflow to be completely customizable and that’s something that I really have to piece together with paperpile right now. Integration with notion will be really nice because I can use their relational database to track my papers, but even their relational database is a little limited tbh

2 Likes
31

Hello, We at Third Iron would love access to a Public API, so that we could integrate with our product, BrowZine and directly export papers to your platform as requested by our customers: https://thirdiron.com/

Please develop this and have a great day!

-Third Iron

32

Hi, still no further info about API? I would like to massively extract cite keys from paperpiles. Is there any way to do it in current version ?

1 Like
33

Yes, you can export your library (Settings>Export). All the export formats (except RIS I think) will include the cite keys.

34

I don’t know if this ever made it to a roadmap for the product, but it should be considered.

There are a number of new tools out there focused around note-taking and organization heavily used in academia I’m sure: Obsidian, Roam Research, Notion, LogSeq, Mem to name just a few. I see more and more tools integrating with Zotero, primarily because they can. And it’s the first time I see people switching back to Zotero simply to have the ability to more easily integrate with all tools facilitating the thought process and the research.

I would take a good look at what the LogSeq and Zotero integration look like for example to get a better understanding of how people hope to use products like Paperpile. Starting around 7:55 of this video: Integrating Logseq and Zotero: A great combination for academic writing - YouTube, you can how, with the /Zotero keyword, the user can connect his notes with a paper and its annotations.

7 Likes
35

THIS! ^ I would love to be able to import my notes into Obsidian.

4 Likes
36

Are there any news on a public API for PP, like Zotero has? Everywhere software is getting more connective and interoperable, why should Paperpile remain a data silo, a lone digital dinosaur?

2 Likes
37

Agreed. There are a lot of features that users want and it would be a lot easier to see these if you were to provide a public API. Users could just make integrations themselves and I bet that would take a load off for developers.

2 Likes
38

This is not about what we want. It’s about to understand what makes costumers stay!

If people work with several connected services, it is more likely, they will keep using it. The more services are connected to other services, the better it is for Paperpile from a economical point of view. Paperpile is the “source of truth” in such a scenario.

I assume Paperpile Devs know this. So the only reasons against a public API I see, are:

  • They are afraid of create an attack surface. As there is already an API, that argument is like “Security by Obscurity”
  • They are afraid of additional bandwidth load and/or are not willing to implement quotas.

Am I missing something?

39

Agreed, where is the API? It’s 2022 and people want to build on platforms.

40

No news to share here yet – although integration with other tools has always been at the core of our strategy, developing a public API is simply not a topic we’ve prioritized thus far. Despite @Alfred_Feldmeyer (and all the rest who have chimed in on this thread)'s valid reasoning, it is a matter of user demand + our available resources rather than lack of interest or potential risks. This remains a long-term goal for us and we keep track of demand.

1 Like
41

It would be very nice to have this be a public api

2 Likes