The impact of data sharing as a scientific practice is ever increasing. Making data available for everyone increases visibility of data creators, allows original results to be reproduced and puts data in a prime position to become a catalyst for novel research. Given the significant costs (time and resources) that are typically associated with obtaining high quality data, promoting increased reuse of data is an economical choice.
Researchers and other professionals that do not have the required resources/connections to produce/obtain their own data sets benefit in particular from repositories like the SMGR, since it provides immediate and convenient access to experimental material that would be hard to acquire by other means.
In the SMGR you will find data revolving around network-forming slime molds. At present this includes images of such networks obtained in controlled wet-lab experiments and equivalent graph representations of said networks.
The best way to find out what's in the SMGR is to check out the contributors page or to skim through the data using our browsing functionality.
SMGR data may be used in any way, provided license agreements and terms of use of individual data sets are observed. This includes reproduction of known results as well as the novel research evolving previously unexplored research questions.
When browsing data sets also look for documents indicating open research questions suggested by the contributors of the sets.
You can expect the data in the SMGR to be of research grade quality. All data is required to be part of some scholarly publication subject to a peer-review process. We thus trust the involved expert reviewers to ensure that the quality of the data is on par with the standards of its field of origin. In addition to that we will inspect data ourselves upon submission.
However, we are not in a position to guarantee that all SMGR data will be perfect always. We still strongly recommend that you carefully inspect any data that you want to use in order to convince yourself of its validity with respect to the intended use.
We are grateful for any support we receive. There are a couple of things that you could do to contribute:
The SMGR is as basic as it gets. We use a standard Apache webserver in combination with a file indexer. Both ingredients we build upon are open source projects.
There are several reasons why we decided to keep the SMGR as simple as possible:
Please note, that the fact that the SMGR only offers the most basic features necessary for repository operation, does not mean that it cannot grow in the future. However, implementation of additional features must be warranted through sufficient community interest.
This is a difficult question with many unknowns. What we can say for sure is that the SMGR will be available until end of June 2022. After that one of two things will happen: Either the current contact person keeps the SMGR running or it gets shut down.
All we can say now is that it is very likely that the next contact person is going to keep the SMGR going. In this case the SMGR will be available for sure until the servers running it cannot be operated any more. Then it will likely transfer in-house to newer hardware and thus continue. Owing to its simplicity maintenance and moving should be relatively easy.
In the unlikely case that the SMGR is forced to stop operation here at the MPI for any reason, all we can do is try to move it out-of-house. In this scenario, we will reach out to find researchers in other departments willing and able to adopt it in order to keep it going. Again the simplicity of the SMGR will ensure that such a step can be realized without running into great technical difficulties.
If the site is temporarily unavailable owing to technical difficulties it is likely that it will be up and running again shortly. Note that all data in the SMGR is part of a larger in-house backup system. Thus, a power-cut or some hardware failing will not affect long term operation of the SMGR. Furthermore, the technical personnel at the MPI does an excellent job at keeping all services provided by the MPI, including the SMGR, healthy and running.
In case the SMGR is shut down permanently, see How long will the SMGR be available?
Follow these simple steps:
At present data sets in the SMGR are not citeable entities. Please cite the scholarly publication the data set is associated with instead. Thus the original contributors are credited. Appropriate information can be found on the contributors page.
In addition to that, you are welcome to cite the SMGR itself in order to give some credit to us. You may use the following BibTex entry:
@article{Dirnberger2016,
Author = {Dirnberger, M. and Mehlhorn, K. and Mehlhorn, T.},
Journal = {submitted},
Title = {Introducing the Slime Mold Graph Repository},
Year = {2016}
}
For reasons of security our servers are configured such that they serve a limited number of requests (per unit time and IP address) and allow only a small number of simultaneously active downloads. While it is unlikely that the traffic exceeds the threshold values, we recommend to start the download again at a later time. It also helps to break the download into smaller chunks if possible.
If the problem still persists after several tries, feel free to contact us.
No, given the current small size and low maintenance of the SMGR, fees are inappropriate.
The operation of the SMGR is supported and funded by Prof. Kurt Mehlhorn and the MPI for Informatics.
Currently the SMGR accepts all data that revolves around networks formed by slime molds. It is important that data is self-contained, well-documented and of potential interest to others.
This includes images of slime molds networks and graphs of the same in particular but is not strictly limited to them. Such data may take various forms and we do not enforce any given file formats as to not restrict the flexibility of the repository. However, we ask contributors to adhere to the standards of their fields.
For some inspiration check out the data currently contained in the repository on the contributors page.
If you are in doubt look at the data which is already in the repository or simply contact us. We are happy to discuss your data and likely find that your doubts were not necessary.
We aim to encourage reuse of existing data resulting in increased credit for original experimental work. To facilitate this, it is paramount that submitted data be well documented and in a format the targeted community commonly uses.
Submitters are encouraged to provide data and descriptive information such that another researcher would be able to evaluate previous findings based on submitted data. This includes raw data as well as any data derived from it.
In general any submission should include the following:
For starters, you will have to decide whether you want us to host your data on our servers, or whether you prefer to host the data yourself. If so, we will put up a link to your server on our contributors page.
In the second case it suffices on our end to provide us with the URL and some relevant descriptive information. Note that data hosted on your servers will not show up in the SMGR data browser. Furthermore, you will also have to provide proper download functionality alongside with the data on your webpage.
In the second case we will provide you with an upload-link that enables you to upload you data on ours servers. Note that the folder structure of the file you upload will be used in the SMGR data browser so please make sure you structured it properly.
As soon as you have decided, feel free to contact us and we will take it from there.
After submission we will verify whether your data contains all the items we expect to receive with a complete submission. If this is the case, we will inspect it and then add it to the available data sets.
Should we find the data lacking with regards to our policies, we will ask you to provide an improved version.
No, given the current small size and low maintenance of the SMGR, fees are inappropriate.
The operation of the SMGR is supported and funded by Prof. Kurt Mehlhorn and the MPI for Informatics.
The answer is, that the SMGR as it stands is simply too small to warrant such efforts. Given the very countable size of the research community interested in slime mold networks, it is probably foolish to expect a large amount of contributors out of the blue. As a result, the number of datasets will likely not exceed large numbers in the near future.
Given these assumptions, features commonly found in larger repositories, such as automatic upload processes, are simply overkill at the moment. However, they may well be added in the future provided appropriate growth of the SMGR and associated community feedback warrant them.
Even large repositories have started small.