Kyle Kingsbury

Good points, and thank you. I have... way too many things on my plate to really negotiate a full merge at the moment, but I'm open to: a.) specific feature...

Sure, go for it.On Dec 15, 2016 20:55, Chris Martin wrote:Am I correct in understanding that this library doesn't support cert revocation? Would it be possible to add a parameter...

I was thinking one could choose logically or time-ordered tags for observed-remove which are amenable to GC. On 01/25/2012 07:09 PM, Reid Draper wrote: > I don't have a suggestion...

Let's say you use snowflake for unique ID generation. Snowflake is k-ordered and has a time component, so each ID can be roughly located in time. Use snowflake IDs for...

It makes sense for certain ephemeral structures (say, purchase orders), and I'm inclined to keep it for completeness' sake. Not really married to the concept, though. On 01/25/2012 07:43 PM,...

Or you could use unix epoch seconds, or use iso8601 timestamps, etc, plus hostnames... there are many possible strategies here. What bothers me is that you might _use_ timestamps for...

I like that idea. I'm cool with implementing a flake-like at the library level. Say default unique tags are [unix_time, host, pid, unique] and for lww-sets, just unix-time. Floats OK...

I would do both; if provided, you use, say "gc_max_items" and "gc_max_seconds" for history pruning. If not provided, it's library dependent. --Kyle On 02/01/2012 08:13 PM, Reid Draper wrote: >...

Yeah, we should use a disk-backed flake time.

Cool!On Jun 15, 2018 07:24, François Royer wrote:It would be great to support Spark, either by wrapping flambo or sparkling and deferring to these libraries. I wrote a short write-up...