Export formats - JSON

[Philip Schuchardt]

I'm trying to replace the current binary format in Cavewhere to use
Metacave format. Currently, Cavewhere uses SQLite and ProtoBuffers to
save and load data. Both of these are binary formats. Although the
format works well, it's not future proof (as many people have iterated
on this list). Metacave format isn't really here to "rule them all",
it's here to replace Cavewhere's binary format. I just wanted to share
something new with people.

Cavewhere is a wizzy program and it saves me time. It uses survex in
the background for loop closure and can even export and import survex
files. Several people use it to convert survex files to compass files.
I've rewritten a survex parser for importing survex data in Cavewhere
and it was a total pain in the ass (and it doesn't fully support the
survex format).

I would gladly port survex's parsing into a library, if give the right
direction. I would require survex to use the library so it stays
maintained.

Phi|ip

On Thu, Jul 7, 2016 at 5:40 AM, Graham Mullan <graham.mullan at coly.org.uk> wrote:
> Data security and future-proofing
>
> There is a well-known maxim in computing that you should never put data into a program if you don't know how you are going to get it out at some unspecified point in the future.
>
> With this in mind, I have little doubt that our successors fifty years from now (if we have any) will not be using any of the tools that we are using now. They will still be wanting to use our data, though, just as I am still using data that was collected in the 1940s and 1950s. If that data has been entered using a simple text editor, as Survex data is, then it will remain wholly available to them. Even Compass data, which is entered using an interface that 'seems' simple becomes less available and useful once you realise that what appears in the data file, when viewed through a text editor is not the native data that was entered originally.
>
> Don't anyone tell me that text editors won't be around in the future.
>
> I still agree with Martin on this
>
> Graham
>
>
>
> --
> Survex http://lists.survex.com/mailman/listinfo/survex