This is just a whole lot of "what if's" rather than anything factual.
There should still be chance to change things before the final implementation of the DAG for new GTLDs is passed so they are what if's... However if there isn't enough people pointing out the failings and the DAG is passed as it currently stands then the issues are very real.
For example
The Registry Registrar split probably one of the most successful policies implemented by ICANN for introducing competition to the domain market and the whole system was simply written out of the new gTLD proposals without any prior community discussion.
A 100,000 limit was inserted "to solve" the issue. Several interested groups were not happy with this change so it was raised recently at the GNSO and they voted to undertake a PDP. The PDP doesn't undo the proposals in the DAG introduced by the ICANN staff to change the model back to the old pre ICANN days of selling direct but if the GNSO finds against it, it makes the DAG proposals that much more questionable.
How much were .com and .net domains in those days, before the Registrar/Registry split? $70?
If ICANN really wants to create competition there are many ways to do it, their current proposals will do very little to create new competition. Its a bit like setting up 500 local football teams and expecting them to not get thrashed if they play in the top professional league.
Haven't you been talking up .info for years? Have you changed your opinion?
I've always liked .info and I think after .com eventually it will end up being being the second most popular gTLD. Why? Because many domains combine with the .info TLD to give good semantic meaning whereas other TLDs have more limited subset of domains which offer the same good semantic meaning when combined with their TLD.
eg berlin.info works better for more uses than berlin.us (I'm sure there are good uses but just not as many)
.com has a huge first mover advantage and it is branded implicitly day in day out by corporate communications. The only chance of changing this is if all the big brands move to .nike and .dell .microsoft and they all no longer advertise .com - But that's a huge if
Such a shift introduces fundamental economic changes to one of the most successful systems in the history of mankind and introduces a whole new set of questions which ICANN has not discussed fully, in public at least.
I think there is still opportunity for new gTLDs to be successful given ICANN is able to create the right environment, but the thinking behind the ICANN DAG is I believe so fundamentally flawed that it will result in far more harm and failed dreams than successes and improvements to the DNS as a whole.
This is why the economic studies requested by the ICANN Board over 3 years ago are so vital to the new gTLD process. There are just way too many unrealistic expectations floating around new GTLDs. Just because a new gTLD exists doesn't mean it will automatically be a success.
Should new gTLDS be allowed to try any way? That depends on how well ICANN designs the framework. The current DAG framework is I believe so flawed the answer is probably not! at least not without a rework of some of the thinking behind the whole DAG process.