What the contemporary account alleged
The surviving WordPress account, published on 10 July 2008, describes an incident on 2 July. It alleges that attackers forced a rapid downward move before attempting the opposite direction. It also reports the operators’ assurance that they regained control.
This is the project’s contemporary account, not a forensic report. We have not recovered server logs, an independent incident investigation or a verified list of affected votes. The article therefore cannot establish the attackers’ identity or quantify the lasting effect on the dataset.
Why a recovered index value does not settle the question
A homepage capture shows what a visitor could read at the moment of capture. It does not prove that every contribution to that display was valid. Equally, a dramatic account of an attack does not tell us which later readings should be discarded. Those decisions require evidence about the actual processing of submissions.
For a researcher, the useful distinction is between an observed display and a validated measurement. Our historical collection deliberately uses the first description.
The lesson for a new voting system
A future index should keep a documented record of exclusions and revisions. If suspicious activity affects a period, readers need to know whether the published number is provisional, corrected or withdrawn. Silent rewriting of history makes independent analysis difficult even when the intention is to remove invalid responses.
Rate limits and bot checks can help reduce abuse, but no control by itself proves that a dataset is clean. Methodological reporting should state how exclusions were made and whether conclusions change under alternative reasonable rules. That is a concrete area where technology partners and researchers could contribute.
Sources & verification
Editorial review: 6 September 2026. Found an error or another archive? Please contact us.