Re: Definitive list of Euphoria libraries?
- Posted by jimcbrown (admin) Feb 12, 2023
- 632 views
I think that Pete doing it all himself is an instance of amazing drive and dedication.
Agreed.
But for OE, to paraphrase your words, it's still 2008.
Again, not mine - https://openeuphoria.org/forum/m/137497.wc
I wonder if the hardest part of being a package system maintainer might be having the ability to work well with folks that one normally would not get along with. After all, that person not only deals with that one's own code, but has to deal with code others wrote.
Oh i agree 100%. It's worse than "not in my back yard", it's "i didn't write it, so it's a no-go at all, period".
Actually, I wouldn't go quite that far. Perhaps this might have been the case back in the closed source days, but since the open sourcing of 3.0 patches and such have easily been accepted into the source. The hard part has been finding volunteers to handle all the steps.
Today it might even be easier. From what I've seen, it's mostly just Greg who has to OK the approach nowadays.
That said, a big part of the problem was myself - I implemented a code freeze on features some time back to get folks wasting time on not-well tested features to instead work on stablizing the code base for a 4.1 release. Instead, these folks mostly left the community and never came back - even after I admitted defeat lifted the freeze with 4.1 still unreleased.
It took a year(?) to get people to realise the original news.e was not running tasks asychronously, despite the obvious that every task always ended in the order they were declared.
Probably due to a lack of volunteers to perform rigorous and timely testing.
Whatever happened to the many website maintainers OE has had
Actually, since the move to the new forum in 2008, we've never had more than two. Only Jeremy and I had direct access to the website on that low level (including the oe.cowgar.com era), then later it was only euphoric and myself, and finally Greg and I.
That said, most of the steps can be done by non maintainers, so there's that.
Basically, one needs to create a ticket, then a branch against the ticket, then fix the documentation in the branch, then regenerate the manuals and stuff in the branch, commit the changes on the branch and push, raise a Pull Request on github, get it approved and merged, and then finally have the new version of the manual deployed to the website.
If it's that much trouble to edit spelling errors on the official OE website, it's no wonder no one will volunteer anything.
I think the bottom line on why things are the way they are, is: people chose this way. They put in place constraints to ensure they get what they want.
Actually, the reason for most of this flow (e.g. insisting on a pull request for peer review) is to avoid issues like the one you found. I believe the typos came about during a time where we experimented with looser controls and less review (in an attempt to make it easier for volunteers to contribute). As you can see, it didn't really work out that well.
That said, an aspect of this does ring true - I added contraints with my code freeze. I learned my lesson from that, though.
Will you find a new volunteer you know nothing about, to follow this through, working with Greg to not mess up whatever he is doing, to fix a spelling error
For most folks who use github to contribute to FOSS, this is actually how one typically does things. The filing of a ticket, managing branches and commits and making and approving pull requests, would all be second nature. Pull Requests and Peer Review also make sure things don't slip in that shouldn't be (in large part helping to deal with the unknown factor of a new volunteer).
Alas, most of the community here isn't familiar with doing things that way, particularly our nondev folks like _tom.
in the online official public-facing wiki??
I guess that's the other thing - the reason this is so hard (relatively speaking) is because it's a typo in the manual, which is autogenerated from the source code and then uploaded to the website. So the version of the manual on the website tracks comments in the source code and is not meant to be independently editable.
If it were just a wiki entry, it would have been fixed long ago.
If they wanted the situation to be different than it is, they'd make sure it's possible.
I think Greg is already working on a new version of the forum and so forth, which might make these things easier. (And nowadays it really is just Greg's call for these sort of things.)
And if someone did chime in and say, "Hey, I'd love to fix this but I'm not sure how to get started, can someone point me in the right direction?" then Greg and others would do everything in their power to make sure the person who chimed in could accomplish it.
And being here over 25 years doesn't make volunteering possible.
Kat
I think these two are more like independent variables. I've been here almost that long but I don't volunteer much anymore due to my expanding professional commitments and some personal life issues. But Greg and petelomax have both been here longer than I and they've done some amazing things single-handedly.
since i was last refused for the job?
Kat
Hmm. Did you ever apply?
Anyways, for interested folks, I'd say to aim lower first - try to become a dev (even a docs only dev) and submit tickets and pull requests and so forth. Much lower barrier to entry and easier to apply.
I wrote three irc.e in attempt to satisfy people, mostly removing features, and none were acceptable.
My memory on this is very blury but I believe that jeremy_c really wanted to get this in, back in the day. So the dev team probably would have said yes to any version that existed, with a few reasonable conditions (primarily no stealing someone else's copyrighted code and permission to modify irc.e afterwards as new issues needed to be addressed or new features wanted). My vague recollection here is that CoJaBo2 denied that these were acceptable and so no version ever made it to the dev team for consideration.
I should make it clear here that CoJaBo2 was just a standard issue forum member (and a minor at the time) and didn't officially have that veto power. So it's a shame that such requests someone went through CoJaBo2 instead of someone actually on the dev team, like myself or DerekP.
Anyways. So today, you'd just need to get Greg's approval. If that code still existed, he might well happily take the patch as-is, and do the ticket/branch/pull request/approval work himself to get it in.
Again, you already know this, but for others - if someone wants Greg to do a new irc.e from scratch - the issue is that without volunteers it's just one item of a long list of things to do. If nothing else (and unfortunately there's a whole lot else), probably JSON, YAML, etc would need to get done first (so modern web stuff would work).
