[trustable-software] An experiment

Niall Dalton niall.dalton at gmail.com
Mon Apr 3 03:46:53 UTC 2017


> Good question; I don't think we have any good way of doing this at the
> moment. Some sort of structured comment in your code which can be
> automatically parsed and linked against tests might work. I'd like it if
> there were a standard for this (and perhaps there is, but I've not seen one
> yet).
>


​Yes; various projects I've been on have had annotations in the code for
issues defining some new piece of functionality, or bug fixes. Systems
weren't well linked (other than some half-assed scripts as pre-commit
hooks).​




> I don't know of anything practical at the moment; Richard Ipsum has been
> working on improving NoteDB, which aims to store review data in git (
> https://public-inbox.org/git/20170324103723.GA7324@salo/T/#u) but I don't
> know if that will be suitable for bug reports. I would stick to current
> industry practice for this, that is, record bugs in an existing, separate
> bug tracker and record the bug number in the commit messages when an issue
> is fixed.
>


​Yeah, ​I think I'll stick to gitlab support for now.





> What else would make iota more trustable than other code? What else would
>> give us confidence we could put this on the internet for an open
>> capture-the-flag competition to break it?
>>
>> What's stopping you from putting it on the internet and organising a
> capture-the-flag straight away? Monitor it and be prepared to pull the
> plug, but getting things out in the open is the best way to find faults.
>


​Nothing stopping it really -- I'll punt until it has enough functionality.​



> More generally, I'm not sure from your email what the goal of Iota is. I
> understand that it's an experiment, but we should try and follow the same
> practice we'd follow for a standard project, and if it were one I would ask
> for a little justification for its existence.
>


Really, there's no justification -- it's just an example that allows me to
tinker with the corner cases on trying to write something 'trustable'. A
distant second is that there's nothing comes to mind to use as a small
programmable way to publish/query sensor readings (whether hw or sw).
​There's lots of big complicated systems, with poor query languages. This
is a small one, with a time-oriented query language.




> Jim
>
> _______________________________________________
> trustable-software mailing list
> trustable-software at lists.trustable.io
> https://lists.trustable.io/cgi-bin/mailman/listinfo/trustable-software
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.trustable.io/pipermail/trustable-software/attachments/20170402/b8496bd8/attachment.html>


More information about the trustable-software mailing list