Show HN: Comments on Science | Hacker News Reader

Show HN: Comments on Science

10 pointsjuretriglav8 comments

Comments on Science is a free and public user-generated database of scientific comments. Commenting on papers in academic journals is a complicated process, with many steps between an idea for a comment and the appearance of that comment in a (mostly closed-access) publication. The goal of Comments on Science is to reduce this friction and enable accessible and free scientific communication.

About the site: I decided to build something useful in the month of May and this is what I made. It has a few features like following (users, papers, or comments), tagging, voting, etc. It was a nice exercise in everything from idea generation to deploying on a web server.

So, what do you think? Do you have any suggestions on what to improve or which features to add or maybe even remove? I welcome all ideas and please don't be hesitant to criticize, input is input is input.

The site is accessible at: http://cos.juretriglav.si

View on HN

I guess this is it as far as HN is concerned. Not bad for a first run! Thank you revorad and knowledgesale for your suggestions! As a result of your advice I'll implement a follow functionality for tags and see what I can do about integrating the Mendeley API in some way.

Even though Comments on Science didn't make it to the front page it was still a very exciting time for me. Today gave me a bit of a boost for future development.

I'll be back, ready for prime time.

Hey I am a grad student in theoretical physics and love both your concept and realization. I would love a little bit more elaborated tag system. Please go on!

Thanks Knowledgesale!

I realize the tagging system needs a bit of work, I think especially a follow functionality for tags would be nice. So for example you could follow "antimatter" and have the new papers tagged with "antimatter" automatically come to your activity inbox. It currently works like that for users, papers and comments. So if you follow a user, if that user posts a new paper or comment, you will get an update in the activity inbox in your profile.

Is that similar to what you had in mind?

Hey, while the subscription feature for new papers sounds like a nice thing, I thought that it can have another useful function.

For example, let us assume that you are starting an exciting new project and need to learn about the most prominent papers in the field of "antimatter". One may expect that there are some classic introductory texts, of very high quality that are obvious to people who have been working in the field for a while but are much less evident to outsiders like you. Googling may not always yield direct recommendation for the best text and you need to consult an expert. There are also some well-known in-depth and concise-but-exact-overview texts you need to know about.

What if there was a voting system on the best review in this specific field? Such a wikipedia of the most important papers in different subfields would be potentially of great use for everyone.

I would love to volunteer for managing of some sections as an experiment if that is of interest to you.

I can see how that would be useful, but I fear it is outside of the context of the site and I'm not sure how it would fit. Comments on Science is supposed to be about comments, not about papers themselves, but I guess the jury's still out on that one. The best I can do, that still fits in the context of the site as it is now, is to have a link where you can list papers tagged with e.g. "antimatter", and order them by score. I'll think about your idea and get back to you on that. I would love to have some help as I am a solo-coder-designer-administrator on the site and it's already proved to be a handful.

Thanks for the feedback! Much appreciated.

Edit: I forgot to add a link for a similar site to that you described, but for computer science: http://jeffhuang.com/best_paper_awards.html Admittedly, the voting is missing, and it's quite narrow in scope, but my guess is sites like that exist for most fields. Have you not found any for theoretical physics?

I moved the site to a new name, it now resides at http://www.tiris.org

Tiris is a play on Thales, the first scientist and Iris, messenger of the gods.

I added follow tags feature, like you suggested.

There is a lot more work to be done, but I think at this point most of it is in getting new users and squashing existent bugs.

Maybe one last thing would be the use of the Mendeley API, but that is really one of the last major features I'm seriously considering.

If people will like and use the service, then I'll talk to them about what features they are missing.

This is such an awesome idea. You should seed the site with lots of papers. Look at the Mendeley API to see if you can leverage it in some way.

Thank you revorad! Seeding the site was my initial idea, as I wanted to have a database of all papers in all publications, that would be just one-click away and you could comment on them instantly.

As I discovered that Google Scholar has no API (I wanted to use it for the initial idea), I switched to user-submitted papers, with data automatically filled in by DOI. There are only a couple of steps required to submit a new paper and I don't have to deal with a lot of different APIs to seed the site (e.g. PubMed, Mendeley).

Also since the site's scoring algorithm is depended on submission time I'm not sure how I would implement scores if my database was pre-filled.

Thanks for your feedback!

Edit: How could I have missed Mendeley API? There is also a competition for the best app, that I think Comments on Science could participate in, if Mendeley API is somehow used. Thanks!

Reply on news.ycombinator.com