Landing page comment: "Write research papers, right inside your browser"
As a researcher, I immediately ask "Why?"
What is wrong with my Latek utils and fabulous MS Word that my research lab gave me for free.
Then, where are some templates for the journals/conferences I could possible use with Authorea. You can reach far more people if you included templates from most journals/conferences. Then I can see this taking off.. Otherwise it is nothing but a fancy text-editor to me with Latek support and is trying to be like arXiv. As it is, arXiv has a lot of noisy publications.
Hey, thanks for the comments. It is true that your research lab gave you MS Word for free, but that doesn't mean someone didn't have to pay for it in order to give it to you. I personally don't like the experience of writing articles in Word, but I can totally see how many people would be fine with it, especially if their work is not equation-heavy.
I prefer writing in Latex more than Word, but it still isn't a happy experience for me. The errors are nearly useless, I'm constantly compiling and waiting for pdf output, and then I have a static pdf as output, which isn't so exciting.
That's why we decided to try and build something different. I want to look at rendered text most of the time during my workflow, but I don't want to use Word. I'd prefer a markup language with a nice-looking, easy to remember, syntax, which is why we offer Markdown. Collaboration should be easy, which is why it's online. I want things to be dynamic, which is why we're working on that. I want to share my data with my figures and my analysis with my data. Again, we're working on that.
It is true that in 2013 what is important is publishing in respected peer-reviewed journals. But we're really trying to think about what an article is going to look like in 2023. I highly doubt it is going to be a static pdf (or static html version of said pdf) available only to paying users of a respected journal, or users whose institutions are nice enough to pay for them.
I'll be the first to admit that we have a long way to go, but we gotta start somewhere.
I second the suggestion to add support for templates for most well-known publishers. For example, in CS even having Springer and IEEE 2-column templates supported would have been enough for me when I was involved in writing papers. I think that when you have publisher template support for the few major publishers in a field, there is much less of a barrier for the researchers in the field to start using authorea. This would IMO not be in conflict with the long term vision of moving papers to web but rather a first step in getting enough users for the publishers to listen.
Thanks for the feedback. This is still work in progress, and we are planning to improve the export function to publication-ready format (as you suggest). This said, we want to bring the authoring process to the web principally for 2 reasons: 1) It makes collaboration easier 2) It expands dramatically the possibilities of sharing scientific content respect to a 'static' PDF format, ultimately impacting transparency and reproducibility of scientific results.
I just want templates for different journals/conferences for the same topic. All their formats are annoying, but at the end of the day, every PI/researchers wants to publish in something that is peer-reviewed. In a university setting, paying to read/download a publication is an after-thought. The entire transparency/reproducibility of scientific results is a good goal but is very long-term goal.
Some entities like ScholarOne/Cambridge Journals wants you to individually upload your images and your text separately. Since a user is just writing their paper on Authorea, maybe, you can keep track of what journals they are aiming for and reformat the publication for multiple journals in ONE SHOT. That'd be the biggest time saver ever.
At the end, doing research is fun but publishing is a painful process, and you are judged based on publications. Reduce the pain.
I totally hear you. And this is part of what we want to accomplish. As for the transparency/reproducibility I do not think this is too much long-term: already now if you prepare a paper on Authorea and publish it on a refereed journal, you can include a link to the Authorea repository. Which means even if the Journal version is static, it links to a 'dynamic' repository which contains the sources of your scientific work. So I believe Authorea is already pretty close to achieving that goal.
What if the Journal asks for a transfer of copyright? Does that violate some kind of law/thing for all the previous (including the final) version of the manuscript stored on Authorea? Would the author have to take down their final-manuscript from Authorea so they can transfer the copyright to the publisher/Journal.
hi there- by and large, authors are allowed to publish the pre-print and post-print versions of their published manuscripts (pre-print being the version right before editor proof-reading, and post-print any version following publication). Authorea articles are essentially "enriched" pre-prints so they do not interfere with the passing of the copyright to publishers.
Something like this is sorely needed. Google Docs is the best compromise between simplicity and collaboration that my group has come up with, but of course it doesn't have reference support.
It is very visually appealing and has good features. However, for me, it seems very sluggish and unresponsive.
This is really neat, a good combination of Markdown & LaTeX.
As one of the founders of https://www.writelatex.com, one of the interesting differences is the fact you're editing the "output" file rather than the source - takes a bit of getting used to! Looking forward to seeing it develop :-)
Thanks John. Authorea founder here. First off, I have tried writelatex for the first time just a couple of weeks ago and it really works great to create beautiful documents. Great work.
Yes, we (and most scholars in general) are very used to editing a source (Latex mostly) and compiling it to something else (a PDF output, or similar). We are trying to streamline that process using the web as the canvas, and render directly to HTML5. In a way you are right, it feels like you are editing directly the HTML output. Which is fine! Our hope is that researchers, especially the ones who already work and write on the web will quickly get used to such a workflow.
The speed definitely needs to be worked on. We were planning on a "Show HN" post once we had modified the architecture a bit. It's all on a single server right now.
Sorry if I exposed you guys prematurely. I stumbled upon it and thought it was a fantastic idea that solves a lot of fundamental issues with collaborating on papers. I've been using Mercurial and bitbucket for this purpose for a while, but it's definitely an issue trying to get collaborators to work this way (if I can even get some of them to clear the LaTex hurdle first). I really hope this becomes a viable option soon.
No apologies needed. The timing is pretty good, we were going to do it in two weeks or so. We've been working on this for about a year now and for the first 8 or 9 months I was telling people that it was usable but not useful. I think it crossed the barrier into useful a few months ago.
I don't know if you've noticed, but we also give the option to write in Markdown. Citations work too, as long as you put them in Latex style, ie \cite{synparb}. Markdown is nearly powerful enough to write a full-featured article, and Latex is clearly overpowered. I like the idea of extending Markdown to give it everything one would need to write an article, without all of the excess.
I've played around with Markdown a bit, especially in writing computatable documents using the IPython notebook. I've actually been hoping to use either Markdown or reST along with pandoc to write a future paper, but I need to do more testing to make sure the conversion is as robust as advertised. But hopefully if Authorea solves the collaboration issue, then I'll have a viable alternative to that approach.
Markdown + pandoc is viable for writing now. I'm doing my thesis with it. Although the citation support gave me the shits so I wrote my own citation markup extension (incomplete - https://github.com/singingfish/Citeproc-Markdown ).
IPython looks amazing. I haven't personally used it yet but it appears to bring most of the Mathematica-notebook goodness over to the python community. The one thing it is lacking, which is admittedly pretty hard to implement, is Mathematica style equation entry. I can input an equation in Mathematica much faster than I can in Matlab or python.
Markdown is a real pleasure to use. One of my favorite features is that it is easy on the eyes. I was never in love with the Latex syntax.
This is excellent. I am writing my phd dissertation right now, and wrote my own variant of markdown to generate it! I wholeheartedly agree! You guys have the right design spirit and vision. Take it forward, and grow a little bolder! Make versions that uniquely take advantage of the web for READING, not just for WRITING!
At the moment it is mostly GitHub-flavored markdown. We did consider Pandoc-flavor, but for the time being we are sticking to GitHub's. We might change it slightly. So far, 80% of our users have authored in LaTeX so we have been busy supporting that
And does your university accept the thesis in the format you output to? The requirements here are quite specific, even LaTeX is nontrivial to set up to fulfill them.
hey- Alberto here (Authorea co-founder). Glad to see us on the frontpage of HN, but as Nate said, we weren't really ready for it. I was cooking dinner! Sorry if we're still a bit sluggish.
We're also trying to make the easy things easy, whether in Latex or Markdown. Drag-and-drop figures is one example. Support for citations directly from ADS is another.
Out of curiosity, can I ask what field you are in?
Computational biophysics - When it's just our lab working on a paper, we tend to use LaTex, but the real headache is when collaborating with experimentalists. They seem deeply invested in Word/Endnote, and it's quite difficult to get them to consider using any other tools (even thought they often hate what they are using).
I've never known anyone in the biological sciences to use LaTex. Everyone here tends to use Word/Endnote because that's what they've always used (and its what the PIs can use). But you're right that nobody likes it. I think tht LaTex is just too complicated for people to get their heads around. Anything that isn't WYSIWYG just isn't too going to fly for the vast majority of biologists. Writing in LaTex is just too much like programming for them. (For good reason - it is programming)
For my last paper though, I was able to do it all with Google Docs, except the citations. I ended up just embedding citations like {author, 2013} and using either Papers or endnote to do the insertion.
However, I still had to submit the article in word format, so I really didn't gain anything and had to effectively reformat it 3 times. I think this has as much to do with Word/Endnote's de facto monopoly as anything. Since the journals require it, we work with it.
All of that said - I'd love to see something like this take off, but there would have to be some buy-in from the publishers.
One thing that really needs to change is that figures need to come to life. While there are a few ways too accomplish this, one is by using javascript libraries such as d3.js to render dynamic images.
[Authorea consultant here] This feature is quite exciting. Using JS based plotting capabilities like D3.js will allow users to do two things at once.
(1) Provide dynamic figures to represent data that is normally ineffective in static form and (2) provide the code and data that was used to create it. This helps others reproduce the results from the data to figure form, which is a feature that is definitely missing in PDF publications.
It uses the sandbox feature for an iframe in html5. It doesn't seem to be supported yet in Firefox. Although if it is I'm sure someone will correct me.
That's a great idea and really good implementation. The only thing I'd miss from Google Docs are comments that are not part of a text (in Docs these are rendered on right side of document). You could implement it by come special syntax for comments, which wouldn't render when exported and on site, and in edit mode, it would show just a little icon. GUI option to click somewhere to add comment in this place would make it even easier. It allows further capabilities, like easy inline peer review.
I'm curious how easy it's going to be to (1) download the git repo from the backend, and (2) format the articles to meet specific journal requirements for submission.
Hi, Authorea co-founder here. A bit surprised to see us on the front page.
1. Full access to the git repository will be added soon. It wasn't our main concern initially as we wanted it to be easy to work on articles for an academic researcher who wasn't familiar with git.
2. This currently requires some post-authorea cleanup. We are also working on making this much easier. The main pain point that we are trying to solve currently is making collaboration on articles easy.
Great to hear that these things are planned. I think #2 is really quite important, because if there isn't a simple path to preparing a submission-ready document, then there is still a major point of friction in the process.
Hi, Authorea consultant here. I completely agree. It's important to provide a seamless way to export Authorea articles to submission-ready format, for example by including the relevant Journal style file (or even giving users the option to chose from a list). Work in progress.
Looks great! One quick comment - the links on the left hand side of an article should only take up enough space to fit the text for the links, so more space can be dedicated to the article.
thanks for the tip! You are right indeed. Not the best use of whitespace there. We are still polishing up the front end a bit. It would be wonderful if you gave Authorea a try for your thesis. Feel free to get in touch for support as rendering a very long document in the browser may be more cumbersome than regular academic papers.
I think some other people mentioned this too, but I'd love the ability to connect this to my own git repository so I can store / have access to all of my changes.
This is a great answer to all this "open up the papers" movement, and going forward a much better solution than those apps that cropped up for asking for a given research paper. If things remain open, that is. A github for papers sounds great, and an actual natural extension of the idea of github. Whereas github is pragmatic projects, authorea is the step before that.
Exactly! And at the same time we can exploit the power of the web, which unfolds into an easier approach to collaboration and a wider range of possibilities in terms of ways to show / plot / share the data and any other products of the scientific research. "Papers" haven't changed at all in the last few hundred years, but research and technology has dramatically. It's time to keep up.
The LaTeX support seems very nice. Nice job! The concept of a paper as a DVCS repository is a very natural one, and it's exactly how I would write my papers back in the day.
I'm curious how the "host the data and the code" comes in. Perhaps you're supposed to clone the backing git repository and add the code to that, but I don't see how to do that yet. Early days.
If you look at a figure in an article, for example https://www.authorea.com/483, you'll notice that the figure reference has been pulled out of the latex source. We did that so that in the online version of an article we can make figures cooler, like attaching data to them. Figures exist in their own folders, and we have big plans for those folders.
Notice though the big conceptual difference between any online (collaborative) LaTeX editor and Authorea. In Authorea articles are written in a browser and live in the browser (as HTML5 pages). There is a huge potential to be exploited by such a paradigm shift.
If I'm writing a paper, it's great for collaboration if it can be written in the browser. Being able to preview my paper in the browser sounds nice too, but to publish it I need to be able to compile it into a PDF.
yes, we (scientists and scholars) still rely on the exchange of PDFs to communicate research. This is a pity. The problem is that a scholarly paper in PDF, even when it is on the web, it is essentially a photograph of a (physical, analog) paper. As such, a PDF fails to expose the living, dynamic, interactive nature of research. The baseline motivation for Authorea is to expose such nature: to expose the data underlying plots and images, to allow articles to be forked and cloned, to enable web-native data visualizations, to get version control right, and so on. We are trying to move in this direction!
I hear what you are yearning for - I remember trying to convince my supervisor that researchers need to look towards github for progress.
His response stuck with me: While scientific research is continuous and never complete, the whole point of publication is that it is a record in a fixed point in time. The results and discussion cannot be altered, or reused: it is a milestone, a historical record of a novel contribution to knowledge. Sure, you might reanalyse your data, rework an image with new parameters what have you, but the nature of sharing your work with your scientific community requires it to be static.
I partly agree with your supervisor. While publications are indeed milestones and are great for obtaining recognition in the scholarly community, they can still exist and accomplish their function alongside their "rich" counterparts: online versions of those publications with live, dynamic, executable data and images, to allow forking, cloning, and interaction. We now produce research in a much more dynamic environment than even just a decade ago. My hope is that 10 years from now a scientific paper "locked" in a PDF format will look obsolete. I may be wrong, but I sincerely hope that.
Indeed. A snapshot could be a tag, and a reference contains that tag. Looking up a reference would let you see the paper in the state that is being referenced, but also allow you to easily move forward to later versions.
We are a long way from that now, but it is appealing.
These two problems are related - I don't mind having the final product in public (it's bound to get there eventually) and thus not having to pay too much (or even use the free tier), but putting everything in public before publication is insane and I doubt it will be used for anything but blog posts for that reason.
hi jclos, I don't think that putting out research content out in the open is insane, as you say. While privacy is indeed a good and a safe idea in a number of contexts, for most research projects, sharing the full sources from the very beginning is only advantageous. For example, less secrecy encourages collaboration, rather than competition. Also, it is a way for authors to claim their ideas and work as soon as they have it. Why wait months before getting something published when we have a fast modern content publication system right in front of our eyes? I have a vision, and I may be wrong. But my vision is to use the web to publish and disseminate scientific content: science you can fork, clone, reproduce, and interact with, from the very first word you jot down in a paper, up to the last equation. My hope is that Authorea is a step in the right direction.
I understand, and I agree to some extent, but the truth is that it is too risky.
In my country at least, but I'm sure it is the same for most, researchers survive by putting their quota (fixed by the ministry of research) of publications in well renowned journals. The risk of having someone "steal" your idea (in the very loose definition) and beat you to the publication, for instance by having a bigger research team, thus putting you out of a job, is a real one.
In an ideal world your system would be perfect, putting the advancement of knowledge before everything else (as open source does with software), but researchers have to look out for themselves before all, considering how their performance is evaluated and how that affects their livelihood. Unlike open source software, there aren't a lot of ways to monetize scientific publications. I would enjoy seeing things change, but change is slow to come. Especially in an area as pachydermic and anchored in traditions as academia.
I really like Authorea and I will use it for non-published work, but I feel that limiting private publications so much is going to limit adoption. But then again what do I know, I guess we will see how it goes.
hi there- well, you could also say that Git is free, while GitHub is not. But that would be a far-fetched statement because GitHub is currently a free service for (hundreds of?) thousands of users who publish source code in the open (Open Source). In a similar vein, Authorea will be ALWAYS free to author and access public, Open Science articles. At Authorea, we like open, transparent, reproducible research and we are trying to encourage people to publish in the open (Open Science). We hope to set up institutional licenses with universities, so that subscriptions will not be a burden on readers or scientists. This is the best model we can think to benefit science and the public, while keeping us alive. Any comments appreciated.
To complete the analogy, if I have a github repository cloned locally, I have no need of github to have full access to the source code and history on my machine. Will I be able to have a local copy of my Authorea paper and all the local tools to construct and view it? Github leverages the distributed nature of git beautifully, and adds value with collaborative features. Will Authorea as well?
Thanks Richard! I am so glad you like Authorea. It was great chatting about Academia.edu last month at the Berkman. I look forward to meeting up soon. --Alberto
Just like, update to the internet, man. I want my articles to look beautiful on the web, not just in antiquated print journals!!!!!
Why, in 2013 do references need to look like this (Johnson-Kines et al., 1533) when we have HYPERLINKS on the web??? And you can show a little snippet in a tooltip--or the actual quote or line of the document that you're referencing?
hi toomim, yes we are still a bit rough around the edges. We are only a few months old. If you can let us know what bugs you found, it'd be great. And we will definitely announce it again as soon as we are ready for a more official launch. Thanks for checking us out
