Durham University

The Durham HepData Project

This site will soon be superseded by the new hepdata.net site, which will be opened for data submissions in the near future.
In the meantime, please continue Submitting Data on this site, then data records will be automatically migrated to the new site.
HepData submission guidelines

Submitting Data to the HepData Database

If you have, or know of, data which you think should be in the database, then please email us at hepdata(at)projects.hepforge.org. The criterium for data to be included is that it be data of a publishable form and not preliminary data. Data from conferences often fall into the latter category. If it is possible that data will change following peer review, it may be better to delay HepData submission until after journal acceptance.

Any text files you send us will be processed into a standard format, which can be seen from the "input" link on any of the existing records. Data submissions provided in text formats closer to the "input" format (not the "plain text" format) can therefore be entered more quickly into the database. You can start by looking at the "input" file for an existing record that is similar to your submitted data. An annotated sample input file is available. Some nomenclature that is still used in practice is defined in the historical 1970s encoding manual (14 MB). It is useful to provide key metadata for each table such as a caption, the process being considered (initial-state and final-state particles), important kinematic cuts, and headers for the x and y columns specifying the quantity with units. Multiple x and y values can be given in the same row separated by semicolons and non-existent entries can be indicated by dashes "-". Numerical values should be given with an appropriate and not excessive number of significant figures (in general, a maximum of 4), avoiding giving more decimal places for the errors than the central values. Keywords should be assigned to each table giving the reactions (processes) and observables.

An online form is available to test data input. Click on "Browse..." to select your input file, followed by "Upload" and "Process", then click "Display" for the first 10 tables, or "Display All" for the whole record. Please note that this online form is only for testing purposes and a record will be overwritten by the next person using the system, unless a valid paper ID and password are specified (see below).

A Python script has been written by Mike Flowerdew (ATLAS) to convert objects in .root files to the HepData "input" format. A test .root file can be made with another Python script and some brief usage notes are provided.

A ROOT TGraph-like class has been written by Christian Holm Christensen (ALICE) to hold results with statistical and systematic errors and with the ability to import from and export to the HepData "input" format.

It is possible to use LaTeX mathematical expressions in the "input" file, which will be rendered using MathJax in the HTML display; see here or here for example records that use LaTeX.

We have now moved to an automated submission system. Administrator accounts, intended to be one per experiment, or one per physics group for large experiments, can be requested by sending the desired username and associated email address to hepdata(at)projects.hepforge.org, then log in. The encoder can then upload data in the "input" format from the online form using the paper ID and password provided by the administrator of their experiment or physics group. Some brief usage notes give more information. A list of existing administrator accounts with their associated email addresses can be found here.

STFCIPPP HepData is funded by the UK STFC and hosted at the Durham IPPP.
HepData also maintains the UK mirror of the PDG.