![]() Since I already use Evernote on a daily basis, it works for me. However, an important quality of any system is that you actually use it. ![]() (This is possible in Notion, but that is something for another post…) However, there are many things I am still missing, such as creating your own fields for each paper, and interacting with the annotations through a spreadsheet. This system has been quite helpful for me with several student projects. No true integration with a reference manager.Limited commenting possibilities (notes from everyone appear the same by default).Could limit the way students explore literature.Everybody can use their own reference manager if they want.Saves time both for me and for students.The system is easy to use, paid account only needed if you want a lot of storage.So somebody with edit permissions would be able to add more of the tags that I use, but not add entirely new tags. It’s good to mention that since the notebook is originally mine, only my tags can be used within the notebook. Still a lot, but now it’s doable to screen the results and narrow them down. So for example if your project is on transfer learning and you want to find all papers I might suggest, the query “tag:ml-transfer & tag:p2” gets you 43 results. With these tags, you can then do queries on topic & priority. p4 – not related to our research but more “general interest”.p2 – important paper for many projects in the lab.p1 – everybody in the lab should read this.I have been using the type, topics and projects for a while, but the priority was an addition after I shared the notebook. Projects (a specific project where I might want to cite this paper). ![]() Topics (specific types of machine learning, applications etc).When I add a paper to this notebook, I add several types of tags: Sharing a collection of 900+ papers is probably not effective □ But what helps here a lot, is the tagging system of Evernote. But you could also choose “can view” option if you prefer. For the students I was supervising, I used the “can edit” as permissions so they could also add new notes, annotate papers etc. Since Evernote allows sharing notebooks, to have a shared collection of papers all you need is to share the notebook with the people involved. Sharing your paper collection with others Since there is no direct link, I might have a paper in one place but not the other, but papers that I cited in my own research in the last few years, are definitely in both. This is my paper collection i Evernote – 913 in total – and each note is a paper (or report, etc).Įach note is at least the PDF I saved (below), and perhaps some notes I made about the paper. The only link between the two is the Bibtex key, which is how I name the note in Evernote. Remember that Evernote is not a reference manager, but it is where I store the paper PDFs and notes about the papers. I was already keeping track of the papers I read in Evernote – see this post on organizing my bibliography with Evernote and Jabref, but I will recap some things here. This might be also an option if you are organizing a journal club. Since I was already an avid Evernote (get 1 month premium for free here) user, I decided to see if shared Evernote notebooks could be the solution to share papers with students. Furthermore, large codebases can be de-cluttered by removing comments not helpful in maintaining code.Although I have only supervised a couple of students during my tenure track, I already found often saying the same thing during each meeting – in particular, what are good papers to start reading about a particular topic. The proposed framework for comment quality evaluation incorporates industry practices and adds significant value to companies wanting to formulate better code commenting strategies. Using neural networks, comments are classified as useful, partially useful, and not useful with precision and recall scores of 86.27% and 86.42%, respectively. Additionally, features based on code and comment correlation are designed to infer whether the comment is also consistent and not superfluous. We develop features to semantically analyze comments to locate concepts related to categories of usefulness. A total of 20,206 comments have been collected from open-source Github projects and annotated with assistance from industry experts. ![]() We conduct surveys and document developers' perceptions on the type of comments that prove useful to maintaining software in the form of comment categories. We propose Comment P r o b e for automated classification and quality evaluation of code comments of C codebases based on how they can help to understand existing code. Approaches to evaluate comments based on whether they increase code comprehensibility for software maintenance tasks are important, but largely missing.
0 Comments
Leave a Reply. |