Kyle Husmann

> I like to use less to view the checkpoint file I do the same thing, but with a lot of free vars the it can be a long distance...

Here's a simple one-liner to make progress/checkpoint monitoring easier: `watch 'cat CHECKPOINTFILE.omx | awk -F "\t" -vOFS=, '\''{print $1,$3,$4,$5,$NF}'\'' | tail | column -s , -t'` output: ``` Every 2.0s:...

@jprindle on which part? my experience is linked in the forums thread

It's linked in the opening comment of this issue: https://openmx.ssri.psu.edu/node/4642

related: another place I expected to get checkpoints with "Always Checkpoint" turned on but didn't is when running mxRefModels -- is this intentional / is there a limitation around this...

More observations with checkpointing: 1) after running a model with checkpoints, if I mxRestore and then continue running the model with checkpoints, it overwrites the checkpoint file rather than appending....

> Yeah, I recommend always using strict=TRUE with mxRestore Ah, good to know, thanks!

> If we change the default, doesn't that break backward compatibility? Sort of? Is there code out there that depends on loading omx files that do not match the name...

> In the present case, it would be good to consider what sort of scripts might get broken, and how easy it would be to repair it. Exactly -- enabling...

Hmmm in `read_resource` is [`field_names`](https://github.com/frictionlessdata/frictionless-r/blob/0881f1c4bbd0c5f679a4f74665e9acd7df0554af/R/read_resource.R#L210) basically the same as [`col_names`](https://github.com/frictionlessdata/frictionless-r/blob/0881f1c4bbd0c5f679a4f74665e9acd7df0554af/R/read_resource.R#L271)? `field_names` looks like it's only used with the col_select assertion at the top there, and is the only thing the...