If you go on step further with double-entry accounting you will need to book the "stack" to the "journal" as required by the tax office. I do not see how this should work with you local-first approach, because you cannot really guarantee the correct order. If something is in the "journal" it has to be "canceled", if the action was wrong beforehand. This is an additional booking step and not a "revert" of the previous action. No conflict resolution will help you here.
Especially the area of double-entry accounting, is something where I never would choose such an local-first/sync approach. Local-first in the classic desktop app single source of thruth sense, sure, but not this way.
And the example #1 isn't even better. Maybe, I am not smart enough, but when the physical clock of user B is _way_ off in the wrong direction, you still get the wrong ordering. Example 1 assumes that the users clocks are going more or less correct, which is sometimes just not the case.
We have an interesting problem here, both target Python and TypeScript with our backend/frontend framework, so we need to generate both Python and TS code from a single logic to generate our programming libraries. This creates a lot of trouble, introducing for example Jinja2 templates to be able to auto-generate TS code from our code base, but this means for us, we have to hand-write those templates without typing in the first place and it is a lot of doubled work.
A typed meta language, like Haxe would be great, but unfortunately Haxe does not have real TypeScript support.
There really isn't a good and feasible solution for this.
Yep, there are bad bugs for example in mod_perl which is written in C and takes the interpreted Perl code and runs it in the Apache context. I think this is what the OP "heard about". But that is not the fault of Perl itself.
On the flip side, freedom is related to independence. If a company would heavily fund the Perl society or another FOSS, it would become dependent and loosing it's freedom. Freedom is something you cannot count or measure with capitalist thinking, but it does not mean there is no value to it.
Perl is not memory safe? Are there pointers directly to memory like in C? No, it is an interpreted language that runs opcode in the Perl virtual machine.
Sure, there are quite some safety concerns with Perl, but they can be mitigated.
For example there is the taint mode with "-T" that prevents direct execution of system commands.
Would I use Perl for a new project? No. :-)
I would be interested in more details about the cmecf hack!?
reply