Re: Could postgres12 support millions of sequences? (like 10 million)

[Rob Sargent]

> On Mar 21, 2020, at 1:13 PM, Peter J. Holzer <hjp-pg...@hjp.at> wrote:
> 
> On 2020-03-21 12:55:33 -0600, Rob Sargent wrote:
>> To me the description of the ID smacks of database-in-the-name folly. I
>> recognize that OP is likely unable to take another path. I’ll not push this 
>> any
>> further. 
> 
> Identifiers often have internal structure. In Austria for example, the
> social security number contains the birth date. Invoice numbers, project
> numbers or court case numbers often contain the year.
> 
> That's because they are used by *people*, and people like their
> identifiers to make some kind of sense. The computer doesn't care.

Since OP said this was digital not paper, I see this as a presentation problem 
bleeding into database design (assuming your response was an invitation to push 
further).