I agree with Shal. I would be very disappointed if Groups.io became limited by what Yahoo Groups was. Bringing groups over from Yahoo groups would provide a short term boost for the site, but in the longer term Groups.io needs to develop into its own service.
One point worth remembering is that Yahoo Groups made the decision to drop the database feature, presumably based on metrics of its use. Now this doesn't automatically mean that this was a good decision - even if only a small number of groups are affected, perhaps these were some of the most important groups on the site.
However, it's worth considering a few points:
- In most cases, the admins maintain the databases, and don't need to provide edit access to normal members. An external service could work well in this scenario.
- Yahoo groups doesn't offer the databases feature any more, so Groups.io doesn't necessarily need to offer this feature to convince people to switch over as long as it is better than Yahoo groups in other ways. Groups.io definitely is better in many ways.
- Wikis provide much of the same features. If Mark adds features like sorting, then this would close most of the gap. Some people have more trouble adapting to technological change than others, but if they need a feature enough, they are likely to learn how to use it.
- Once the feature is added, it needs to be maintained. Further, removing a feature that has been added will annoy a large number of users and reduce the trust of user's who don't even use that feature because a feature that they need may be removed in the future. Therefore, it seems that if a database feature were to be added, it should only be added when it becomes clear that such a feature is necessary. The Wiki will work well for most small groups, larger groups are more likely to need the specific functions of the database. Since there aren't large groups on Groups.io, this need isn't clear.
- A generic API would solve this use case and so many others.
- Re the alternative of spicing up the Wiki: The ability to sort tables in a Wiki is cool, even there was a separate database feature as well, so the effort of going down this path wouldn't be wasted.
For all these reasons, I'd strongly recommend against implementing a database feature at a current time.