What would replace Stormont petition of concern?
Arlene Foster's indication that she would like to see the Stormont petition of concern scrapped has focused attention on issues like same-sex marriage.
Its introduction has so far been blocked by the veto power despite attracting the support of a narrow majority of MLAs.
However, no-one should leap to the assumption that the Stormont parties will drop the controversial petition system in isolation, without revisiting all of the rest of the Good Friday Agreement's "ugly scaffolding".
Back in 2014, a Stormont committee reviewed the petition of concern system and concluded that there was no consensus for reform.
The committee considered whether the use of such petitions should be restricted to certain key areas.
It also discussed whether petitions should be triggered by weighted majorities of 65%. Again it could not achieve consensus.
Then last year the Alliance party tried to stipulate reforms to the petition of concern system as a precondition for taking the justice department again in a DUP and Sinn Féin led coalition.
The former Alliance leader David Ford claimed Arlene Foster thumped the table in exasperation at this suggestion.
The cross-community voting system, which is triggered by petitions of concern, was introduced as a guarantee against majority rule in Northern Ireland.
Nationalists had long argued that the border was an effective gerrymander in order to guarantee a unionist majority. So they were never going to be prepared to participate in an assembly which might look like a recreation of the old pre-Troubles Stormont parliament.
If the petitions of concern were dropped in isolation, then same-sex marriage might progress. But equally there would be nothing to stop, for example, a unionist majority changing, say, the definition of a victim.
Some argue that the system should be changed so cross-community voting can only apply to constitutional or Troubles related matters.
But some politicians will baulk at any attempt to tightly define what they might regard as a "key area".
As the DUP leader put it, she thinks her critics want to keep the veto power for what they think is important but prevent her party deploying it in relation to its priorities.
If the current cross-community system of "parallel consent" (a majority of both unionists and nationalists) is dropped that would logically mean there is no need to get MLAs to sign in as Unionists, Nationalists, or Others when they take their seats.
One argument in favour of this is that it might enable Stormont politics to evolve, with a vote for smaller parties or others appearing less of a "wasted vote".
But there is still the problem of "majority rule" - which means most commentators have talked about requiring weighted majorities in order to ensure some level of cross-community backing for controversial policies.
However, this in turn could lead to dilemmas. The TUV, for example, has suggested a weighted majority of 60%. This threshold would have blocked the November 2015 vote in favour of same-sex marriage just as assuredly as the old petition of concern. Moreover a coalition of unionists and others could, theoretically, have outvoted the nationalists who made up just 37% of the seats in the old assembly.
The Stormont committee's 65% suggestion would - using the old assembly break down - require some nationalist buy-in, but not the assent of Sinn Féin. However, at this threshold the old DUP team would have still been able to block initiatives they opposed.
The arch critic of the Stormont system, Bob McCartney, once told me that the working of the assembly was like a Heath Robinson contraption and that any attempt to alter it only added an extra layer of bizarre complexity.
So it's not just about abolishing the petition of concern, but working out what might replace it.
Certainly the post-election talks are unlikely to achieve any breakthrough on these basic building blocks of the Good Friday Agreement in a matter of weeks.