The first news is that with only 28 seats, the DUP, even with the support of the TUV's Jim Allister, can't use the PoC; the 30 vote effective veto that each community (nationalist/unionist) has over legislation. The translation of this is that gay marriage could be coming to Northern Ireland.
As you can see, compared to past elections, this one had very high turnout. Additionally, there are a few important things to note.
First, as with 2016, 2011, and so on, this election was really 3 concurrent elections.
First, the Unionist election.
In this, the UUP failed just about as hard as it is to fail. While the party did make gains on the DUP, the reality is that fighting a party so scandal prone, the UUP should have taken 3 votes for every 2 they obtained. The UUP has been on a downward trend, and continues on said trend, and frankly, may well find itself dead in a decade or two. Remember that the UUP was "the" party of Northern Ireland for many decades. It may simply be the party has too much baggage and history. Unless things turn around (and frankly, they blew this, their best opportunity so far to do so) the UUP will vanish and be replaced within the Unionist electorate by another Unionist party.
The TUV meanwhile actually lost votes. This despite the strong 'drain the swamp' message (both Belfast and Washington DC were built on former swamplands) and despite being a clear and strong anti-nationalist alternative for the scandal ridden DUP. Allister, again, remains the only elected MLA and when he decides to resign, it is likely his party will go with him.
The DUP comes out as the winner of this, but only just. While they did generally maintain their massive lead, taking roughly 2/3rds of the unionist vote, they also fared very poorly compared to the nationalist and other vote. The big problem is they became transfer toxic. 28 seats is below the threshold for petition of concern, and only just ahead of SF.
The Nationalist election was more joyful.
The SDLP, while they fell behind, losing ground to Sinn Fein, managed to effectively gain seats. 12 seats in a 90 MLA assembly is a gain on 12 in a 108 MLA assembly. It is no secret the SDLP benefited from UUP transfers. This is also the first time the SDLP won more seats than the UUP. What is important is that unionist voters now have experience transferring to the SDLP. Even with a push against it, it is likely the SDLP will remain a final ballot spot for many of these voters seeking to block SF.
PBP, not officially a nationalist party (but still enjoying the support and transfers of a subset of nationalist voters) also lost some support, dropping to 1 MLA.
SF is one of the big winners of this election. Their push to become the #1 party, while it failed to achieve their goal, certainly built momentum behind the party. Additionally, the entire Nationalist electorate did amazingly well, losing only 2 seats, compared to 16 for Unionist parties. This puts SF in the driver's seat for the negotiations that will be coming over the next few weeks and months.
The Others also gained
The Alliance and Greens both managed to hold on to their seats, and non-sectarian parties in general did well. Naomi Long in particular had a very strong showing in East Belfast, electing a second MLA in the seat. The party also did fairly well in South Down, important as the current Alliance is very much a Belfast centered party, winning all of their seats inside or beside Belfast. Expanding outside this area is an important step in growing the party.
Interestingly, this would give the DUP and SF 3 executive (cabinet) seats, and both SDLP and the UUP 1. However, due to the way these seats are calculated, should either the SDLP or UUP withdraw and remain in opposition, the Alliance would then qualify for a seat. Should both withdraw but the Alliance not, the DUP would qualify for 4, and should all 3 remain in opposition, the DUP and SF would both qualify for 4. The latter scenario could present problems as any desire to keep Sugden as Justice Minister would mean someone (either the DUP or SF) would have to donate a seat, and unlike last time when the balance was 5-3, giving up a seat this time means giving up a majority on the executive.
That all assumes this assembly will function. I personally have strong doubts. This is a re-aligning election, as was the 2003 election. In 2003, the lead party in both electorates changed, while in 2017 what we've seen is that "Unionism" is no longer the mainstream default option for voters, who are more likely to cast a first preference for Others and Nationalists, and are more willing to transfer to both these groups.
My current thinking is that this assembly will never sit. Negotiations will likely take most of the year, and we will likely see another election within 2 years.
3 years 3 elections; 50% chance of happening
This would see NI return to the polls at around this time next year after the 5 major parties have come to an agreement on how to proceed.
2019 election; 25% chance of happening
This would see the next assembly election take place at the same time as local elections in NI. This is most likely to happen if negotiations are not fully complete, but all 5 parties agree they are far enough along that an election is useful and helpful.
Delayed Assembly; 15% chance of happening
This would see the current assembly sit, but not within the current timeframe set out in the GFA for such a thing to happen. This could happen if, in 6 months time, the parties come to an agreement and also agree to use the assembly elected in this election.
Breakthrough; 10% chance of happening
This would see a government in place by the end of April, and the newly elected assembly sitting without interruption.