Questions that are asked frequently about ecoCompute Science, together with a number of points that are worth clarifying in advance.
No. ecoCompute Science is a new peer-reviewed, artifact-first conference held online. The ecoCompute conference was a large in-person engineering conference, held in Munich in 2024 and in Berlin in 2025. An in-person event of that size takes a lot of organising, so there will be no in-person ecoCompute in 2027. The earlier material remains available in the archive.
No. It is not a research-only event. Submissions from companies, public bodies, open source projects and independent practitioners are as welcome as those from universities. They go through the same track and the same review. What every submission needs is a tangible artifact that reviewers can run, use or inspect.
Yes. A product, an internal tool, a dashboard or a production system is a good artifact, as long as reviewers can use it and check what the write-up says about it. A hosted system is fine if reviewers have access for the whole review period. For data that cannot leave the company, see the question on data that cannot be published below.
Summer 2027, online, in the European time zone. There is no venue. Exact dates will be announced on this site.
Two pages and a tangible artifact that reviewers run, use or inspect. See the Call for Papers for the format and Artifacts for what the artifact must contain.
Yes. The write-up is a guide to the artifact, not the evidence itself. Bullet points are acceptable throughout and presentational quality is not a criterion for acceptance. The full paper follows later, at stage two, after the event.
No. An archive file or a container image is sufficient. What is required is a precise, step-by-step guide that enables a reviewer to verify the reported results without consulting the authors.
No. A tool, a dashboard, a notebook, a spreadsheet, or a dataset together with the code that produced it are all acceptable. What matters is that the artifact supports the points the write-up makes, and that the guide argues how. Every number in the write-up still needs a source: an output of the artifact, the data included with it, a cited publication, or a stated assumption. See Artifacts.
Yes, provided reviewers can be given access to those systems for the duration of the review period. Hardware requirements should be declared at submission so that reviewers able to work with them can be assigned.
By stating which property is expected to be stable and within what tolerance: an ordering, a direction or an effect size, together with the variance across your own runs rather than a single mean. A reviewer who reproduces the stated ordering but not the absolute values has reproduced the claim, provided the claim was formulated in those terms.
That is the purpose of the kick-the-tires phase. The opening days of the review period are reserved for setup: reviewers report any issue that prevents them from starting, and authors are given a short window in which to correct it. Artifacts should not fail on setup problems.
Yes, and more of them would be welcome: the optimisation that did not yield a benefit, the effect that disappeared once noise was controlled for, the tool that did not measure what it claimed to measure. An artifact-first event is the right place for such results.
Yes. An unsuccessful attempt to reproduce a published result is a full submission here, not secondary material.
Yes, if the data can be anonymised while preserving the properties on which the claim depends. Scaling every value by the same undisclosed factor, for example, preserves orderings and relative effect sizes. The guide states what was changed and which properties are preserved. See Artifacts. Where no such anonymisation is possible, as with some industry data that cannot leave the company, the work can still be submitted with another tangible artifact, for example the design of the system, the calculation behind the claimed effect, or the method applied to a public configuration. The claim is then limited to what that artifact supports. Authors with such work are invited to contact the organisers.
No. The event consists entirely of discussion, following the session format used at ICT4S. Each accepted submission is allocated three sessions, the author participates in all three, and all participants have read the submissions one week in advance.
Yes. The submissions are circulated one week beforehand and having read them is a condition of participation. Groups are limited in size so that every participant is able to contribute, which requires that everyone arrives prepared.
In two stages. Stage one is the two-page write-up together with the artifact, reviewed and presented at the event. Stage two is a full paper written afterwards and shepherded by a committee member, which receives a DOI. Stage two is published in ECEASST, the Electronic Communications of the EASST. It is a diamond open access journal hosted by Berlin Universities Publishing, properly indexed, with no fee for authors or readers.
Undecided. Artifacts are difficult to anonymise reliably, so a decision is required between accepting single-blind review and adopting a fully open process. Comments on this question are welcome.
Undecided. Registration free of charge produces high rates of non-attendance, which is untenable for discussion groups of limited size. A nominal fee, a short application, and an invitation model are all under consideration.
Yes. A programme committee and an artifact evaluation committee are currently being assembled. Please write to us, stating which kinds of artifacts you are able to run, use or inspect and what hardware you have access to.
English.