3 ms·
Hi everyone! I'm Dawid, the Founder and CEO of Bootstrap Shuffle during the day and a reader interested in science at any other time. According to the study[1
by kemyd 8y ago
Hi everyone!
I'm Dawid, the Founder and CEO of Bootstrap Shuffle during the day and a reader interested in science at any other time.
According to the study[1], emotional people (I am one of them) overestimate the knowledge they acquire by reading just the titles of articles shared on Facebook.
[1] https://journals.sagepub.com/doi/10.1177/2053168018816189 https://journals.sagepub.com/doi/10.1177/2053168018816189
Bullets.tech is a response to it. This is a place where we publish summaries of articles for science enthusiasts. Five bullet points or less. The very essence of the article.
If you ever happened to share or like a science-related article after reading just the title, you will like our website :)
If the title of an article says that "a cure for cancer has been found", our summary will allow you to find out that the article is referring to one of the hundreds of tumor types and not in humans but in mice.
You can browse Bullets.tech:
1/ via the https://bullets.tech https://bullets.tech website
2/ By signing up to the weekly newsletter on the website
3/ Using the terminal https://www.npmjs.com/package/bullets-tech-cli https://www.npmjs.com/package/bullets-tech-cli
4/ on Facebook Group https://www.facebook.com/groups/bullets.tech/ https://www.facebook.com/groups/bullets.tech/
5/ and the RSS feed: https://bullets.tech/rss/latest.xml https://bullets.tech/rss/latest.xml
I am willing to collaborate with people with similar interests (science, technology, psychology, cosmos, history, medicine, nature) who would like to take part in developing this project as non-technical co-founders.
PS. I copied my post from Product Hunt because it explain a lot (i think)
Any questions? Maybe suggestions?
- Iv 8y agoThank you for this, that's really great! You are totally right: between the full article and the (often over-hyped) title, there is a sweet spot for a short, high signal-to-noise summary. Even the abstract is often too verbose or omit some important shortcomings of the article. In some subreddits, users receive a lot of karma to do things similar to what you are doing. Usually making: - a summary of the finding (the abstract in one sentence) - a summary of the methodology (which often tells if a study is solid or not) - context to understand if this is a breakthrough, an interesting but unpractical effect, a potentially misleading discovery These are especially precious in environmental or medical research where one study may seem to say totally different things based on the overall context.
- vezycash 8y agoHi. Thanks for this. I'd love it if the "load more" button can work without javascript. A paginated link would we wonderful.
- rokhayakebe 8y agoCan we get this for Tech and possibly other verticals?
- kemyd 8y agoFor now, we’ll publish articles from various categories in one feed (science, technology, psychology, cosmos, history, medicine, nature - as mentioned above)
- janekm 8y agoI like the idea, and am certainly going to keep it on my reading list. I would really appreciate some indication of the strength of the evidence in a paper though... Some combination of the quality of the methodology, statistical relevance (including sample size and size of effect), whether it falls in line with existing evidence...
- vitaminCPP 8y agoGreat idea! I want this feature in all news paper.
- stringlytyped 8y agoA browser extension that shows the Bullets.tech summary when you browse to an article that has been summarized would be very useful.
- jedimastert 8y agoSo these are all hand-summarized? Are there any guidelines or standards for summarization (what kind of information would be most useful, what wouldn't normally be included, how are articles found?), and if so, would you consider putting them on an "about us" page? Also, who are the summarizers? Hand picked? Volunteers? I'm not assuming any kind of ill-will or mal-intent, but how do I know that the bullet points will be any more useful that the title in terms of bias, other than goodwill? I really really like the idea, and even the execution, so these are definitely just kind of nit-picks.
- kemyd 8y agoYes, we are summarizing articles "manually". You raised a valid point (bias) and to be honest, I don't have an answer right now. We will create about us/FAQ page for sure. I need at least a few hours to think about it.
- ar-jan 8y agoSuggestion: also add a link to the original research article (not just the newspaper article or press release etc. that's being summarized). This would be very helpful.
- edibleEnergy 8y agoHeya, I really like it :) Like the top comment currently though I have a hard time with the font. Everything else seems great!