(I am not subscribed to this list, sorry for messing up the threading)
Hi,
I'm the original founder and implementer of the sigrok project, back in
2009. I have not been involved in a long time, though I still like to
keep an eye on things, like this list.
It's been painful to see the sigrok project die over such a long time.
When I felt I could no longer be maintainer, I did the honorable (and
normal) thing: I passed the buck to my then co-maintainer, Uwe. When he
could no longer be maintainer, he did the same, and chose Soeren as the
new maintainer.
Soeren lost interest a long, long time ago, but did _not_ do the right
thing and find a new maintainer. I've been hoping somebody would step in
and either take over, or fork the project. As Ladislav said, it's an
important piece of tooling for many people.
I feel I must correct a few things.
Soeren Apel wrote:
> he two core issues for me are that libsigrok's architecture is
> fundamentally outdated and that it is *impossible* for me to review
> and merge changes to libsigrok device drivers.
You are wrong on both counts. The libsigrok architecture is great, as
shown by the fact that it effortlessly supports hundreds of devices
across tons of device types. The problem is rather that you don't
understand it, by your own admission, and thus don't feel able to review
code for it.
Not being able to review the code is no excuse: I've offered my help,
and reviewed nearly every libsigrok PR on github, and held your hand
through pushing many of them through. This was last year. I eventually
stopped trying when you stopped responding to mails. One set of drivers
I asked you *three* times to push, and you ignored it all.
> On the architecture side, it would make way more sense to spend time
> on rebuilding libsigrok so people can write device drivers in
> languages other than C - python or rust, for example.
We've talked about this, and it simply isn't true. Writing a device
driver is easy, and there's lots of examples and tooling to help. It's
no coincidence that so many people have written drivers.
If you want to see more device drivers, all you have to do is *merge*
the ones people have been submitting for years, not rewrite the
foundation they're based on. The problem is you, not libsigrok.
> That's why libsigrokflow was born to have a gstreamer-like data
> pipeline that is as flexible and powerful as can be. Unfortunately,
> we never got very far because
This was always a terrible, terrible idea, and I said so at the time.
Not only is it not necessary, it's such an obvious case of second-system
syndrome -- very much a utopian ideal of smashing together all notions
of drivers, in/out modules etc into a language agnostic thing that would
solve all of your problems. Predictably, it came to nothing. I can't
*believe* you're still mentioning this failed idea SIX years after even
the repo has been touched. Don't you think it's time to face facts?
> - People who can perform PR reviews on drivers and give me a thumbs up
> or down on whether to merge or not.
I did *exactly* that, and you stopped responding and merging.
- Someone who is willing to spend some time with me to conceptualize a
> way forward in terms of gradual changes to the project architecture
> and document it on the wiki.
I made several suggestions just last year, and you ignored them completely.
Soeren, you need to finally, after much too long, step down and give
somebody else a chance to maintain the sigrok project. Your inaction
over so many years is incredibly corrosive to community enthusiasm for
the project, and has done a *lot* of damage. Please do the right thing now.
I hope somebody else will step up and offer to maintain, or co-maintain
(ideally) the project, and that Soeren will honorably hand over the reins.
As for me, while I am unable to go back to being a maintainer, I remain
willing to help where I can. I reviewed dozens of libsigrok PRs, and can
do so again. If people step in, I'll be there to help with architecture,
coding, reviewing, the wiki, and everything else that needs doing. I
just don't wanna be the guy on top, is all.
--
Bert Vermeulen
[email protected]
_______________________________________________
sigrok-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sigrok-devel
