I think it is interesting to consider the opportunities data sharing could have with qualitative datasets. Considering the length of time qualitative data collection can take in comparison to a lot of numerical methods, qualitative data sets are often small, limiting their usefulness. Furthermore, these datasets can be used often benefit from reanalysis by other researchers, be that with the same or different methods, considering the nature of most qualitative data analysis. I think data sharing would massively facilitate topic-specific qualitative dataset size whilst also increasing how much can be learnt from each piece of data originally, which is very exciting!

I work with a lot of genetic data, and the genomes I get from collaborators is often of much higher quality than the ones freely available on PubMed or GenBank; what I mean to say is that you need to be very careful about how much you trust free data online. Just because it was published in GenBank does not mean the reads are of great quality or don't need any FastQC quality analysis. In fact, often times they are the worst quality reads in my dataset. This is partly because they are older and therefore from sequencing platforms that may not be as up-to-date. It's still a fantastic resource and I wish more people would upload the genomes they collect online. Just be sure to "clean" your downloaded genomes!

Sharing research data could be dangerous in some context. In a university, in which you work as a graduate research assistant, but also are involved as co-author in many research works, how you manage your data represents how trustworthy you are in front of your supervisors eyes.

One of the challenges I've encountered around the sharing of research data is a lack of understanding among researchers as to what constitutes data. Generally, STEM researchers (and some in the social sciences) have a clear understanding of what constitutes the data underpinning their research, and are well served by institutional and funder policies about open sharing of data, research data management, and so on. However, current policies are, in my experience, unhelpful or unclear when it comes to research data for humanities researchers. It's often a real struggle for those of us working in research support to reach humanities researchers -- sometimes they don't perceive their research as involving 'data', or current policies are unhelpful when it comes to management and sharing of humanities research data (which can often just be primary and secondary literature, and their own notes).

Data from the mobile health app involved in my research can be used to generate a heat map of dominant areas in a particular location that food marketing is having an influence on young people. The current data from this project is limited but further input from participants and in reaching areas further afield could strengthen the use of this data. This data could then be used by other researchers to link certain food ads with particular behaviours and lifestyles of young people in Ireland. It could also be used for ongoing research to support campaigns or public health initiatives that aim to reduce such influences and promote healthier behaviours and lifestyles in an attempt to halt the obesity epidemic.

I am currently involved in the development of a phd and, as usual, we have been generating products in different aspects. In our case, programming code, methodologies to integrate information obtained from different portals on the internet, such as uniprot, hgnc, among others, have produced knowledge bases that integrate all this information to describe a genetic regulatory network. In this case all the objects of the network, that have been considered, have associated knowledge trees that describe their identity, their biological processes, molecular functions, and at the same time,  knowledge bases that model protein-protein interactions. Now, I can see how each one of these different products can be shared immediately, without having to wait for the doctorate to end. This opens a possibility that sincerely I had not matured enough and I am very happy about this philosophy in which all the different products of a research can be considered as shareable. This opens  doors to those of us who are approaching for first time this Open Science initiative.

I am a Research administrator and funding advisor, when means that I support researchers when writing proposals. The Research Data Management team, lead by the Library. The advice is that a DMP should be prepared in the proposal phase even if not required (e.g. in the Research Council of Norway the DMP is required in contract preparation phase). There is resistance to use time to prepare a DMP during the proposal preparation phase as there are so many other Things to prepare/Write. I Wonder how to approach this at the proposal phase and what it the normal practice in other Universities.

I collect plasma parameters and proteomic data from animals right now (and I relate both). A problem I have is to confirm that each "batch" of expeiments are comparable to other batches. to minimize this I include internal controls. Another issue that I often have are missing values, but this is not frequently reported in papers, while it happens in reallity.  

We have had to think about this recently, regarding our research. After performing the test we have developed, we have a lot of information in the form of numeric CT values that indicate the level of methylation on certain genes, which then indicates the likelyhood that a particular patient has cancer. the clinicians who will use this data have indicated that this format is overly complicated for them and does not help them to form a diagnosis. We are exploring different ways of presenting the data so that all people who use it can understand it.