FMS releases Total Access Analyzer and Total Access Speller

This comment is related to the post about Huron from 31 October 08. That post is already closed for new comments so I am continuing here. I was wondering how it would work for a scenario in which different users have different privileges to access different data. I assume for this scenario that all users will have a front end that is linked to a backend. An administrator would have no problems and would just synchronize all backend and front end tables and objects to his local machine. But a user with lesser privileges would have to also synchronize all tables since his front end will not open if not all linked tables exist in his backend. So maybe you would say that this user will have the table definition in his backend somehow, but that this table’s data will not be updated and synchronized from the cloud. But how is he going to have this table in the first place, if it is not available to him on initial first time downloading from the cloud?

Another option may be that this user will have all tables synchronized regularly, but that his front end will prevent him from accessing some data from his backend. I think this is not a good option, security wise, because the backend can be accessed directly not via the front end. Another problem with this scenario is that according to the video the user is the one that selects the tables to synchronize. This can cause confusion. Some uses will not know which tables to select, and this seems to me to be error prone. So maybe there should be a way to use code in order to synchronize, and that this code should allow synchronization of the table structure only, without synchronizing the data it contains. I hope I made some sense.

Thanks Gilad

Hi Gilad, As of right now, we have not fully implemented the security model for data access priviledges and we are certainly looking for input like this on how to best implement it for Access. So far I have heard all kinds of different requirements from being able to partition data at a table, column and row level. I do see your point about requiring at least the linked tables to exist and I would suggest that this is something that would be controlled by the publisher. In your question, "how would he have the table in the first place", i think this is pretty easy to accomplish as we put the entire .accdb file (including the table schema up to the cloud). The only thing that does not exist is the data from this .accdb file which is synchronized up to actual cloud database entities. As such, if your talbes need these linked tables they would already exist in this .accdb file. Sure maybe you chose to only sync 5 of the 10 tables but all 10 tables would still exist in this .accdb file so I think you will be fine. To your other question about publishing data, I do not expect that a general user will do this. I think it will be more of an administrator (with different permissions). The only think the user will have is the ability to subscribe to what has already been published. They will not need to know which tables to select as this is already part of the publication. They would just need to know to be able to select the right publication. Also to your point about using code, I also agree with this because whenever you provide a solution like this there will always be cases where it does not solve everyones problems. It is for that reason that we do want to provide the underlying sync framework components that power this Huron sync experience so that a developer can use it to customize the experience for their own needs. Certainly this will be a little more work on their part then the point-and-click capabilities we have here but it also gives them a whole lot more flexibility to customize the solution. I hope that helps. Liam Cavanagh