[ProgClub programming] top: st : time stolen from this vm by the hypervisor
roland at rolandturner.com
Thu Jun 6 09:45:06 AEST 2019
On 6/6/19 6:13 am, John Elliot V | ProgClub wrote:
> On 5/6/19 12:35 pm, Roland Turner via programming wrote:
>> If you're struggling to produce reliable systems at your current level
>> of complexity then you'll fail completely with larger, more complex
>> systems; even well-tested ones. There are habits of mind required that
>> you're not yet exhibiting.
> I've been trying to figure out the implications of this, and I think
> perhaps I am just a small systems kind of a guy... there are plenty of
> small systems out there, and far fewer more complex ones, so hopefully
> there is a place for little old me...
No doubt, but you missed the point. You appeared to be advancing the
idea that reliability is somehow a function of large, well-resourced
systems. This is false. Reliability is _*much easier*_ in small systems.
If you're not capable of building reliable systems, then you're not
in any realistic sense a competent programmer.
1: and as before, not that this is not a binary condition; absent
numeric performance targets it's a statement about the path, that at
each step you are making choices which choose more reliable over less
reliable, rather than refusing any effort towards reliability on the
basis that it won't be perfect
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the programming