Time |
Nick |
Message |
00:39 |
MTDiscord |
<wsor4035> something i think we should address at some point is nesting stuff rather than a flat mess. for example the three translation pages could be under /translation |
00:51 |
MTDiscord |
<greenxenith> https://github.com/minetest/dev.luanti.org/issues/19 |
00:52 |
MTDiscord |
<wsor4035> you should edit that to be discord/irc so that certain people dont scream and make that the sole focal point |
00:52 |
MTDiscord |
<greenxenith> I do not care in any way shape or form |
00:52 |
MTDiscord |
<greenxenith> A majority of it was not discussed on IRC |
00:53 |
MTDiscord |
<greenxenith> If anyone whines about it, it will be marked as offtopic |
00:53 |
MTDiscord |
<wsor4035> 🤷 |
00:59 |
MTDiscord |
<wsor4035> all the incoming mapgen stuff should probably be shoved under /mapgen or something |
00:59 |
MTDiscord |
<wsor4035> https://cdn.discordapp.com/attachments/926231483155378176/1317657187992080404/image.png?ex=675f7b62&is=675e29e2&hm=859b272e10b652fb35ab127653b66c88082bc59905607b634838f61e0340eee7& |
01:04 |
MTDiscord |
<greenxenith> Technically premature to close up shop on minetest_docs since we haven't actually agreed to merge it yet but I won't complain about encouraging the merge :p |
01:12 |
MTDiscord |
<wsor4035> merging https://github.com/minetest/dev.luanti.org/pull/20 in 5 |
01:12 |
MTDiscord |
<wsor4035> trival |
01:24 |
MTDiscord |
<wsor4035> pr for this https://github.com/minetest/dev.luanti.org/pull/21 |
01:25 |
MTDiscord |
<wsor4035> merging in 10 |
01:58 |
MTDiscord |
<mark.wiemer> My first PR for the new wiki site 🙂 https://github.com/minetest/dev.luanti.org/pull/22 |
01:59 |
MTDiscord |
<mark.wiemer> You mean if anyone tries to push to the master branch? Or does the current rules prevent people from creating too many remote branches? |
02:06 |
MTDiscord |
<wsor4035> it means that only the master branch can exist |
02:06 |
MTDiscord |
<wsor4035> ideally everyone uses there own repo for branches |
02:07 |
MTDiscord |
<wsor4035> (there is no rules) |
02:07 |
MTDiscord |
<wsor4035> other than informal announce here |
02:43 |
MTDiscord |
<mark.wiemer> https://github.com/minetest/dev.luanti.org/pull/23 🙂 |
02:44 |
MTDiscord |
<wsor4035> how is that redundant? |
03:11 |
MTDiscord |
<mark.wiemer> Because the left side panel is clearly visible already :/ |
03:12 |
MTDiscord |
<wsor4035> https://github.com/minetest/dev.luanti.org/pull/23#pullrequestreview-2504287422 |
03:13 |
MTDiscord |
<wsor4035> its mainly just a drop in for a table of contents that existed there on the old site. hence just burn the whole section |
03:14 |
MTDiscord |
<mark.wiemer> This line specifically feels out of place, it's got a typo, not capitalized, no ending period, and only points to the navigation bar, which is pretty standard for websites nowadays. The other part of this section, the link to terminology is valuable |
03:15 |
MTDiscord |
<wsor4035> before you start, look here ......then do nothing |
03:15 |
MTDiscord |
<mark.wiemer> Ah, maybe we can rephrase that section to something else. But for now that line is bothering me :/ |
03:16 |
MTDiscord |
<wsor4035> i mean, it would be ideal to have it all make sense before merging, but given the whole repo content is a trashfire. 🤷 just going to merge it |
03:21 |
MTDiscord |
<wsor4035> for https://github.com/minetest/dev.luanti.org/issues/15 should we stage the stuff in a branch, or just commit it as the work is done? only possible reason i can think of for branch staging is seo/duplicate content policy maybe? |
03:21 |
MTDiscord |
<wsor4035> does the current wiki support redirects? if so could add redirects there once the stuff is merged in |
03:23 |
MTDiscord |
<mark.wiemer> I'd vote to merge pages as they're valuable, easier to review things in smaller chunks. Maybe as we marge pages we can also clean up typos, formatting, etc. a bit |
03:24 |
MTDiscord |
<wsor4035> > Maybe as we marge pages we can also clean up typos, formatting, etc. a bit no thanks. one for one import, than fixes, whatever can be done |
03:25 |
MTDiscord |
<wsor4035> small stuff whatever |
03:25 |
MTDiscord |
<wsor4035> but if you doing batch imports of stuff mixing is with changes is a bad idea for history, turns it from a straight forward review to a debate over content changes, etc |
03:26 |
MTDiscord |
<greenxenith> Yeah, cleanup should be separate prs |
03:28 |
MTDiscord |
<wsor4035> seems you can only do inner wiki redirects as a user https://www.mediawiki.org/wiki/Help:Redirects |
03:30 |
MTDiscord |
<wsor4035> could do it in https://wiki.luanti.org/Distributing_Minetest_Games style i suppose and just hope googles crawler ai stuff figures it out |
03:31 |
MTDiscord |
<wsor4035> but also i really dont want to have to deal with the pain of touching the wiki |
05:00 |
|
MTDiscord joined #minetest-docs |
08:56 |
MTDiscord |
<rollerozxa> I can take care of doing manual "page moved to dev wiki, click here" redirects for the wiki as things are moved over. for SEO it should make Google be able to determine it as a new canonical URL since it is under the same domain and the original URL becomes uninteresting to index |
09:08 |
MTDiscord |
<rollerozxa> merging imported pages as we convert them sounds like a good idea. I don't think we should wait with it or keep it in a separate branch, but merge it so that it becomes gradually available on dev.luanti.org as we go |
09:13 |
MTDiscord |
<rollerozxa> hopefully we won't end up with revert wars with one of the wiki busybodies... but they seem to have more or less abandoned it at this point |
10:03 |
MTDiscord |
<rollerozxa> ah great, I can't create new branches in a fork when I have write permissions on the repo |
10:10 |
MTDiscord |
<rollerozxa> got a workflow going anyways |
11:08 |
MTDiscord |
<greenxenith> You can't make a new branch on your own fork? Does forking copy branch protection? |
11:46 |
MTDiscord |
<rollerozxa> I can |
11:47 |
MTDiscord |
<rollerozxa> I just have to go to my fork specifically because github won't redirect me to it if I have write permissions |
13:34 |
MTDiscord |
<rollerozxa> merged a trivial PR, removing some [toc] markers that came from the voxelmanip wiki |
13:35 |
MTDiscord |
<rollerozxa> merging PRs 25 through 29 in ~10m (I checked so that the commands in 25 were correct now that I'm at my computer) |
13:35 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317847495409270834/image.png?ex=67602c9f&is=675edb1f&hm=1cdb63e82717ca66d43cac7dd01b8b362b7377644fb700b7a6a0fec3bd665ff9& |
13:37 |
MTDiscord |
<luatic> there are further things that could be said on the luajit PRs |
13:37 |
MTDiscord |
<rollerozxa> there probably are |
13:37 |
MTDiscord |
<luatic> do we need to open a PR for everything or can we just shovel stuff in? |
13:37 |
MTDiscord |
<rollerozxa> but this is just importing it from the voxelmanip wiki |
13:37 |
MTDiscord |
<rollerozxa> well I think we restricted pushing to the master branch |
13:38 |
MTDiscord |
<rollerozxa> I don't know how formal we're going to be with the  PR merge process |
13:39 |
MTDiscord |
<luatic> i think for many changes mandatory PRs are just complicating the editing process, making it slightly worse than a wiki even |
13:40 |
MTDiscord |
<rollerozxa> yeah trivial changes could just be pushed to master without any PRs |
13:40 |
MTDiscord |
<rollerozxa> like https://github.com/minetest/dev.luanti.org/commit/ef715c78ff41755653d6b0dc6bdeda5098bea112 |
13:40 |
MTDiscord |
<luatic> not even just trivial changes, also more elaborate changes, wiki style |
13:41 |
MTDiscord |
<rollerozxa> well of course! |
13:41 |
MTDiscord |
<rollerozxa> if we were to take the wiki approach to heart we would commit everything to master and then do follow ups if there are further changes |
13:41 |
MTDiscord |
<luatic> i get strict approval rules for a piece of C++ software that will blow up if someone dereferences a null pointer, but for docs it's not that critical |
13:41 |
MTDiscord |
<luatic> i genuinely think that wouldn't be a bad idea |
13:41 |
MTDiscord |
<rollerozxa> yeah |
13:41 |
MTDiscord |
<rollerozxa> at least for things we in the docs team are sure about |
13:42 |
MTDiscord |
<luatic> especially since there's a lot of work to be done and reviews - which are quite the chore - will slow everything down significantly |
13:43 |
MTDiscord |
<luatic> the documentation here is also notably supplementary and not authoritative in nature (yet), doc/* in the luanti repo still has the last word |
13:43 |
MTDiscord |
<rollerozxa> indeed |
13:44 |
MTDiscord |
<rollerozxa> also I was looking for something resembling admonitions for hugo and found https://github.com/martignoni/hugo-notice |
13:45 |
MTDiscord |
<mark.wiemer> I agree with removing the PR requirement for trusted contributors, that's how we do it at my job too 🙂 |
13:46 |
MTDiscord |
<mark.wiemer> GitHub repo > settings > rulesets > bypass list  Should get you what you need |
13:47 |
MTDiscord |
<rollerozxa> well, they sure stick out |
13:47 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317850464037109841/image.png?ex=67602f63&is=675edde3&hm=34139926e871579984c1ac87705acfcbf37fb06ec51cfce52c036f21bbb0cb98& |
13:47 |
MTDiscord |
<rollerozxa> because there's nothing built into hugo for doing this? |
13:48 |
MTDiscord |
<rollerozxa> and with the light theme |
13:48 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317850740001210459/image.png?ex=67602fa4&is=675ede24&hm=c5ac424ce732505bedd8596fd72511381cedb8d4ecd6e159902acc5f5d67ed97& |
13:48 |
MTDiscord |
<mark.wiemer> I'm pretty good with CSS, happy to look at how Hugo works and add some custom styles as needed. I'm a big fan of keeping things lightweight. In Markdown you can include most HTML tags |
13:48 |
MTDiscord |
<mark.wiemer> I hadn't heard of Hugo before yesterday but I'm familiar with the concept of a static site generator |
13:49 |
MTDiscord |
<luatic> as always the light theme is better 💪 |
13:50 |
MTDiscord |
<rollerozxa> I'm more of a jekyll boy but I've heard of hugo before and it seems to be quite nice just from what I've worked with it right now |
13:50 |
MTDiscord |
<rollerozxa> as long as we have the shortcodes being used throughout the docs we can switch out the implementation and styling of it afterwards |
13:50 |
MTDiscord |
<rollerozxa> Y/N? |
13:51 |
MTDiscord |
<luatic> yeah i wonder, are custom HTML tags not an option? |
13:51 |
MTDiscord |
<rollerozxa> shortcodes are like includes with custom HTML behind it |
13:52 |
MTDiscord |
<luatic> why can't we use custom HTML directly |
13:53 |
MTDiscord |
<rollerozxa> because shortcodes are shorter? |
13:53 |
MTDiscord |
<luatic> this does not look very short to me |
13:53 |
MTDiscord |
<mark.wiemer> What is a custom HTML tag? I'm just talking about span, div, p, etc |
13:53 |
MTDiscord |
<luatic> and HTML should be more portable |
13:54 |
MTDiscord |
<rollerozxa> are you kidding me? |
13:54 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317852193185075210/image.png?ex=676030ff&is=675edf7f&hm=0d74204a06acbc61f8814653ff4581ed5c0014b0415aea735a237e3527b2e298& |
13:54 |
MTDiscord |
<mark.wiemer> Lol I can't tell if luatic is joking 😬 |
13:54 |
MTDiscord |
<rollerozxa> me neither |
13:54 |
MTDiscord |
<luatic> Mark Wiemer: HTML5 allows custom HTML tags. I suppose it depends on the Markdown implementation whether they survive. |
13:54 |
MTDiscord |
<mark.wiemer> Oh dang Ive never heard of that! What's the value? |
13:55 |
MTDiscord |
<mark.wiemer> Yeah hence my desire for custom code lol, there's gotta be a better way |
13:55 |
MTDiscord |
<luatic> Something like <my-note>something to note</my-note> should be possible |
13:55 |
MTDiscord |
<mark.wiemer> Given the SVG implementation, my vote is no |
13:55 |
MTDiscord |
<rollerozxa> I literally linked the better way which will be above, luatic wants us to do it like below |
13:55 |
MTDiscord |
<luatic> no |
13:56 |
MTDiscord |
<luatic> option 1, without custom html tags: <div class = "my-note">note this thing</div> option 2, with custom html tags: <my-note>something to note</my-note> |
13:56 |
MTDiscord |
<rollerozxa> should I remove the vector icons? |
13:56 |
MTDiscord |
<mark.wiemer> This is basically just automatically applying a set of attributes to an existing tag, I assume? Feels like too much niche work when global CSS already exists and should cover everything we need for a wiki site |
13:56 |
MTDiscord |
<rollerozxa> option 2 is basically what hugo shortcodes are |
13:57 |
MTDiscord |
<luatic> i have not yet checked in detail how we would have to configure hugo to make either of these work, but both of them are in my opinion better than  {{< notice note >}} {{< /notice >}}  both in terms of verbosity and portability |
13:58 |
MTDiscord |
<luatic> (and all of these are a downgrade from AsciiDoc admonitions but well, can't have everything) |
13:58 |
MTDiscord |
<rollerozxa> please, we're not forking hugo |
13:58 |
MTDiscord |
<mark.wiemer> Yeah. Custom HTML tags are fine, the question is whether enough people know about them. Obv they de-dupe things which never hurts. But anything with curly braces feels awkward in HTML/MD |
13:59 |
MTDiscord |
<mark.wiemer> Ah, didn't see the SVG was just for the icon. Honestly that's not too concerning to me, somehow I thought the SVG was the whole background lol. I am fine with adding that library. SVGs are verbose but contained and don't concern me when used properly. And this looks like a proper use to me |
14:00 |
MTDiscord |
<rollerozxa> would you like it to use bitmap icons instead? |
14:00 |
MTDiscord |
<luatic> have we considered gfm alerts? |
14:00 |
MTDiscord |
<mark.wiemer> Um.. no? Â Gfm = GitHub flavored Markdown? |
14:00 |
MTDiscord |
<luatic> yes |
14:00 |
MTDiscord |
<rollerozxa> that was literally what I was searching for and this was the closest I could find |
14:01 |
MTDiscord |
<mark.wiemer> Sry the "um no" was for roller's bitmap suggestion lol |
14:02 |
MTDiscord |
<luatic> https://github.com/CodeYourFuture/curriculum/pull/887/files#diff-060012ebaf8eb6a1bc35b4547fc84a240d8b68789be6fc65267d52e0f36080bbL16 |
14:02 |
MTDiscord |
<mark.wiemer> If Hugo supports it GFM alerts would be good, I just hope we can customize them a bit as needed. Worst case we run a macro later to change to a new standard  I vote whatever is simplest for now, then we can evaluate later. For me the core priority is actually moving the docs, not finalizing an admonition implementation  Gonna walk the dog now 🙂 |
14:02 |
MTDiscord |
<luatic> these people are apparently moving away from the thing you found to GFM alerts now that goldmark (and hence hugo) supports them |
14:03 |
MTDiscord |
<luatic> not the thing you found per se, but a similar shortcode |
14:03 |
MTDiscord |
<luatic> personally i can live with gfm alerts despite their obvious inferiority to the admonition syntax i came up with when i was 14 or something |
14:04 |
MTDiscord |
<rollerozxa> well that's good, exactly what I was looking for |
14:07 |
MTDiscord |
<rollerozxa> (I don't really understand your distaste for vector icons then) |
14:16 |
MTDiscord |
<wsor4035> Regarding custom html, css. That's a hard no fir maintaince, ease of use, etc. Would have written my own static site generator if that was the case |
14:17 |
MTDiscord |
<rollerozxa> yeah i would prefer using an off the shelf plugin/theme for this |
14:17 |
MTDiscord |
<rollerozxa> because it seems that now I'll have to write my own CSS for this or bring in the CSS from the notice theme |
14:18 |
MTDiscord |
<wsor4035> Are gfm notices not supported? |
14:19 |
MTDiscord |
<rollerozxa> hugo allows you to hook them, but you bring your own layout and design of it |
14:20 |
MTDiscord |
<wsor4035> Heh. I say just use the plug in you found and it will be easy enough to replace them on the future if Hugo starts styling them |
14:22 |
MTDiscord |
<rollerozxa> yeah. I think doing a regex find and replace to convert the notice shortcodes to the corresponding GFM alert syntax in bulk would be possible if we were to switch it out in the future |
14:22 |
MTDiscord |
<wsor4035> To do so you will need to turn on unsafe markdown parsing, so probably should fix what that complains about currently |
14:22 |
MTDiscord |
<wsor4035> I just ignored it till now |
14:22 |
MTDiscord |
<rollerozxa> hm yes I noticed that when I ran it locally |
14:24 |
MTDiscord |
<rollerozxa> https://github.com/alex-shpak/hugo-book/tree/7c78a39c531aa2492ed7e92f2ce9dfb2c8c0d3fa?tab=readme-ov-file#shortcodes |
14:24 |
MTDiscord |
<rollerozxa> > It is recommended to set markup.goldmark.renderer.unsafe=true if you encounter problems. |
14:25 |
MTDiscord |
<wsor4035> You have to for shortcodes it seems |
14:25 |
MTDiscord |
<luatic> just don't use shortcodes |
14:26 |
MTDiscord |
<luatic> at least not for this |
14:26 |
MTDiscord |
<rollerozxa> strange because the notices rendered fine without it |
14:34 |
MTDiscord |
<mark.wiemer> Sorry, I have no distaste for vector icons, I actually love them. I was originally complaining because I thought the HTML you shared used SVG for all the styling (header color, background color, etc) instead of using CSS where appropriate. I was wrong. Vector icons are good 🙂 |
14:35 |
MTDiscord |
<mark.wiemer> I feel like if we have to enable "unsafe" settings to get something to work, we should probably just use a different thing |
14:44 |
MTDiscord |
<luatic> I don't see the big problem with a need for "custom" CSS. Surely there's something that can just be grabbed off the web. |
14:47 |
MTDiscord |
<rollerozxa> we do not need to enable unsafe to get the notice theme to work |
14:47 |
MTDiscord |
<rollerozxa> (and I don't know why the default shortcodes in the book theme needs unsafe) |
15:14 |
MTDiscord |
<rollerozxa> anyways so I thought I could import the markdown converted contents of minetest_docs from the voxelmanip wiki: https://github.com/minetest/dev.luanti.org/pull/31 |
15:17 |
MTDiscord |
<rollerozxa> after that I can fix up the asciidoc admonitions/alerts/notices/whatever you call them that were plaintext in the markdown conversion to be proper notices again |
15:27 |
|
erle joined #minetest-docs |
16:16 |
MTDiscord |
<wsor4035> the one file your missing is already in markdown https://github.com/minetest/minetest_docs/commit/1f6d56da06641474396d34f6a5f5827e53625b4d |
16:16 |
MTDiscord |
<rollerozxa> yup, I saw that |
16:17 |
MTDiscord |
<wsor4035> also those probably should be nested under /docs or something |
16:17 |
MTDiscord |
<wsor4035> also ci is erroring |
16:18 |
MTDiscord |
<rollerozxa> oops |
16:18 |
MTDiscord |
<rollerozxa> didn't commit everything |
16:19 |
MTDiscord |
<rollerozxa> hugo server gives me a false sense of security because for some reason it can keep limping on even if there are page errors |
16:20 |
MTDiscord |
<wsor4035> since you already made them live, probably will need to alias the current pages |
16:32 |
MTDiscord |
<rollerozxa> I wonder if we could move some other stuff into nested folders too |
16:33 |
MTDiscord |
<rollerozxa> but I'm a bit unsure how to split things up |
16:33 |
MTDiscord |
<rollerozxa> for minetest_docs the classes were in their own folder called classes |
16:33 |
MTDiscord |
<wsor4035> yes, i said as much yesterday somewhere |
16:34 |
MTDiscord |
<wsor4035> in the mapgen import pr im just taking care of nesting at the same time |
17:02 |
MTDiscord |
<wsor4035> anyone want to take a look at https://github.com/minetest/dev.luanti.org/pull/24 real quick? |
17:11 |
MTDiscord |
<rollerozxa> was just about to say I had skimmed it through and it looks okay 😄 |
17:11 |
MTDiscord |
<wsor4035> lol, if you want to "redirect" the pages on the wiki when you have time |
17:11 |
MTDiscord |
<wsor4035> since your volenteered |
17:12 |
MTDiscord |
<rollerozxa> oh, yes |
17:14 |
MTDiscord |
<rollerozxa> hmm, we're not gonna do anything with the translated wiki pages right |
17:14 |
MTDiscord |
<wsor4035> correct, there generally even worse than the english ones |
17:15 |
MTDiscord |
<rollerozxa> yeah... |
17:15 |
MTDiscord |
<wsor4035> hugo and the template does support translation. i checked when picking them |
17:15 |
MTDiscord |
<wsor4035> other than that idc. some translator can figure it out |
17:22 |
MTDiscord |
<rollerozxa> I wonder, how should we do casing for the page paths |
17:22 |
MTDiscord |
<rollerozxa> that's probably something important to figure out before it's too late |
17:22 |
MTDiscord |
<wsor4035> yeah...... ive just been matching live for now |
17:23 |
MTDiscord |
<wsor4035> lowercase is the generally accepted answer, while using - or _ for spaces is debated |
17:26 |
MTDiscord |
<rollerozxa> I do feel mediawiki is quite unique in how it uses underscore for URLs |
17:26 |
MTDiscord |
<rollerozxa> dashes feel more standard for URLs |
17:37 |
MTDiscord |
<rollerozxa> hmm why does ordering break when I set titles for pages in the front matter |
17:40 |
MTDiscord |
<rollerozxa> guess we'll have to set titles for all pages to make that not happen |
17:44 |
MTDiscord |
<rollerozxa> but anyway, if we have the page /Timing_and_Event_loop from minetest_docs, I can then move it to /docs/timing-and-event-loop/ and make an alias? |
17:49 |
MTDiscord |
<mark.wiemer> I'm in favor of kebab-case, feels most common in URLs, snake_case works too |
17:50 |
MTDiscord |
<rollerozxa> yeah I would want to use kebab-case too |
17:51 |
MTDiscord |
<rollerozxa> I'll go with it unless someone objects |
17:52 |
MTDiscord |
<mark.wiemer> Did we make a GitHub issue for this for tracking? I'm thinking we can also add a rule to enforce proper front matter |
17:52 |
MTDiscord |
<mark.wiemer> How do folks feel about enforcing CI to pass before merging any PRs? Too early? |
17:52 |
MTDiscord |
<rollerozxa> like something hugo would enforce or in a page style guide? |
17:53 |
MTDiscord |
<mark.wiemer> Both, ideally an automated enforcement before PRs are completed, with an informative error message. Not sure if it's built in to Hugo or would need to be added some other way |
17:53 |
MTDiscord |
<rollerozxa> I would hope there is some way to do that |
17:54 |
MTDiscord |
<rollerozxa> I also wonder if there is any way to deduplicate the initial markdown header line when the frontmatter title is defined without needing to modify the book theme |
17:54 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317912690961944627/image.png?ex=67606957&is=675f17d7&hm=3949d129786de1330170d1a489188abbf36e26900973b1ffb73410fae988ee29& |
17:54 |
MTDiscord |
<rollerozxa> just a bit of a minor inconvenience though |
17:57 |
MTDiscord |
<mark.wiemer> Nah I think title is what shows in the tab and heading 1 is what shows in the page. I'd live with it, we're pushing into premature optimization at this point I'd say |
17:59 |
MTDiscord |
<rollerozxa> I would still want them to be consistent though |
18:00 |
MTDiscord |
<rollerozxa> but that's for if someone wants to standardise the casing on the actual page titles... I'll lowercase all the paths so things at least won't move around then |
18:01 |
MTDiscord |
<mark.wiemer> - https://github.com/minetest/dev.luanti.org/issues/40 Â - https://github.com/minetest/dev.luanti.org/issues/41 |
18:02 |
MTDiscord |
<mark.wiemer> I think the filenames convention should match the URL convention for simplicity, right, maybe just less encoding? So the filename would be docs/my-article.md |
18:04 |
MTDiscord |
<mark.wiemer> https://github.com/minetest/dev.luanti.org/issues/42 |
18:04 |
MTDiscord |
<rollerozxa> yeah, the filenames in the repo match the URLs they get generated to, so I'm renaming all of the files and adding titles and aliases to the front matter |
18:05 |
MTDiscord |
<mark.wiemer> Aliases for backwards-compatibility to dev.luanti.org or for some other reason? Isn't this site super new and WIP, or am I missing something? I guess I'm worried the aliases might be more clutter than they're worth |
18:05 |
MTDiscord |
<rollerozxa> we've imported stuff from the old developer wiki and dev.minetest.net redirects to dev.luanti.org |
18:05 |
MTDiscord |
<mark.wiemer> Ah gotcha, makes sense. Aliases are a go from me 🙂 |
18:38 |
MTDiscord |
<rollerozxa> that was quite a lot: https://github.com/minetest/dev.luanti.org/pull/43 |
18:39 |
MTDiscord |
<rollerozxa> not all aliases for the pages are necessary, and I will likely cull some of them for pages I imported from the voxelmanip wiki as I can just redirect those to the proper URLs when I clean things up there |
18:42 |
MTDiscord |
<rollerozxa> I just went ahead and merged it. if there are more aliases necessary we can add those as 404's get discovered |
18:43 |
MTDiscord |
<rollerozxa> but I think this should be a good standard going forward |
18:49 |
MTDiscord |
<wsor4035> did you pull in http api yet roller? |
18:51 |
MTDiscord |
<rollerozxa> not yet, will do that |
18:51 |
MTDiscord |
<wsor4035> now that the url formatting is done, https://github.com/minetest/dev.luanti.org/blob/master/hugo.toml#L6 maybe can be removed? |
18:52 |
MTDiscord |
<wsor4035> not srue if that will affect aliases or not |
18:54 |
MTDiscord |
<wsor4035> given that the two projects are merged for better or worse. wonder if as triager i should transfer over some of the issues that are worth saving and cull the rest? |
18:55 |
MTDiscord |
<wsor4035> (minetest docs and dev wiki) |
18:56 |
MTDiscord |
<rollerozxa> sounds good |
18:56 |
MTDiscord |
<rollerozxa> huh, what does that do |
18:56 |
MTDiscord |
<rollerozxa> > Set this to true if you use capital letters in file names |
18:56 |
MTDiscord |
<wsor4035> it allowed the shitty Random-casing_of-Titles.md |
18:57 |
MTDiscord |
<wsor4035> if it doesnt break the aliases at all, can remove it |
19:25 |
MTDiscord |
<rollerozxa> honestly I'm confused what such a setting would do for a static site generator since it can't really influence how the web server will serve the pages |
19:25 |
MTDiscord |
<rollerozxa> but let's see what happens if I remove it locally |
19:26 |
MTDiscord |
<wsor4035> it changes the output file names |
19:26 |
MTDiscord |
<wsor4035> so filename Random-casing_of-Titles.md becomes /random-casing_of-titles |
19:27 |
MTDiscord |
<wsor4035> rather than /Random-casing_of-Titles |
19:29 |
MTDiscord |
<rollerozxa> err |
19:29 |
MTDiscord |
<rollerozxa> I see. okay, yeah it doesn't affect alias redirects |
20:04 |
MTDiscord |
<wsor4035> done |
20:04 |
MTDiscord |
<wsor4035> should i just make a commit to the minetest_docs readme that says see dev.luanti.org/docs? |
20:10 |
MTDiscord |
<grorp> so if you have a simple edit, you may just use the edit button and do it on the main branch? |
20:11 |
MTDiscord |
<wsor4035> sure |
20:11 |
MTDiscord |
<grorp> nice |
20:12 |
MTDiscord |
<wsor4035> probably just say your editing something here on the off chance someone else is doing something 🤷 |
20:15 |
MTDiscord |
<wsor4035> i wish i could tell github to just keep my forks master up to date with the forks master all the time |
20:22 |
MTDiscord |
<wsor4035> merging https://github.com/minetest/dev.luanti.org/pull/51 in 5 minutes. @rollerozxa have fun |
20:23 |
MTDiscord |
<rollerozxa> don't forget the titles in the front matter |
20:23 |
MTDiscord |
<rollerozxa> (also redirect time! :D) |
20:24 |
MTDiscord |
<wsor4035> whats the point of that? |
20:25 |
MTDiscord |
<rollerozxa> it will make the pages stay ordered in the page list |
20:25 |
MTDiscord |
<rollerozxa> (and for some pages it fixes the title it got in the pagelist, but for those pages in the PR it's just about having it be ordered properly) |
20:26 |
MTDiscord |
<rollerozxa> I honestly don't know why the reordering happens though |
20:26 |
MTDiscord |
<wsor4035> fixed |
20:26 |
MTDiscord |
<rollerozxa> looks good now, merge time |
20:27 |
MTDiscord |
<wsor4035> since your around, any thoughts? |
20:27 |
MTDiscord |
<rollerozxa> sounds good |
20:27 |
MTDiscord |
<rollerozxa> I assume this is where we will be maintaining it from now on |
20:27 |
MTDiscord |
<wsor4035> minetest_docs repo content has been moved to https://github.com/minetest/dev.luanti.org/tree/master/content/docs |
20:27 |
MTDiscord |
<rollerozxa> I mean I've already started by updating the SecureRandom docs 8) |
20:29 |
MTDiscord |
<wsor4035> done |
20:32 |
MTDiscord |
<rollerozxa> very satisfying getting to clear out a bunch of wiki pages with devwiki links, reminds me of the good old wiki cleanup days |
20:38 |
MTDiscord |
<wsor4035> thoughts on adding bookCollapseSection = true to our _index.md files? |
20:39 |
MTDiscord |
<rollerozxa> hmm, what does that do? |
20:40 |
MTDiscord |
<wsor4035> allows the nested section to be togglable (closed by default) |
20:40 |
MTDiscord |
<rollerozxa> sounds like a good idea |
20:45 |
MTDiscord |
<wsor4035> done |
20:50 |
MTDiscord |
<rollerozxa> looks good |
20:52 |
MTDiscord |
<rollerozxa> https://cdn.discordapp.com/attachments/926231483155378176/1317957479245549568/image.png?ex=6760930d&is=675f418d&hm=5564fd07fa798dde8d8ce2737f4f573d31d22d8e5fc1ceb90d20dda21fe47e67& |
20:52 |
MTDiscord |
<rollerozxa> oops. well, that's what happens if you forget the front matter |
21:13 |
MTDiscord |
<wsor4035> merging https://github.com/minetest/dev.luanti.org/pull/57 in 10 |
21:14 |
MTDiscord |
<rollerozxa> lgtm |
21:38 |
MTDiscord |
<grorp> changing content/engine/_index.md to fix dead links |
21:58 |
MTDiscord |
<grorp> https://dev.luanti.org/building/ is mtg-specific: It describes mtg nodes and behavior, even outside its mtg section. It also mixes up nodes/blocks. In general I'm not sure who's the target audience of that page. |
21:59 |
MTDiscord |
<wsor4035> part of wiki migration, pulling anything that has some level of potential value. then it can be redirected, burned, etc |
22:00 |
MTDiscord |
<grorp> And with a page called "Building", I'd expect "how to compile", not "how to place nodes" |
22:00 |
MTDiscord |
<rollerozxa> we're trying to move things player related that isn't explicitly MTG related to this wiki so that it can eventually replace the regular wiki too |
22:00 |
MTDiscord |
<rollerozxa> (the page should probably be called placing though) |
22:01 |
MTDiscord |
<wsor4035> https://github.com/minetest/dev.luanti.org/issues/15, https://github.com/minetest/dev.luanti.org/issues/12, see dev irc discussions, zughy forum post, etc |
22:05 |
MTDiscord |
<grorp> I wonder if anything of value would be left in that page without the mtg stuff, and whether a "how to place nodes" page is useful to anyone |
22:06 |
MTDiscord |
<grorp> could be useful if contained in a "how to get started" guide |
22:06 |
MTDiscord |
<rollerozxa> that's a good point to be honest |
22:06 |
MTDiscord |
<rollerozxa> that whole page should probably stay on the regular wiki for now |
22:20 |
MTDiscord |
<rollerozxa> now that the accessing android data page has been moved off the regular wiki I can finally expand the instructions to include some images |
22:21 |
MTDiscord |
<rollerozxa> I'm really happy with how the new wiki is coming along |
22:24 |
MTDiscord |
<greenxenith> So are we agreeing to merge the minetest_docs project |
22:31 |
MTDiscord |
<greenxenith> Thats what it seems like we are doing but it doesn't look like anyone formally supported the proposal or anything |
22:35 |
MTDiscord |
<greenxenith> Unless I am misunderstanding #43 |
22:36 |
MTDiscord |
<wsor4035> Sure. Complaints can be forwarded to the period of inactivity |
22:38 |
MTDiscord |
<greenxenith> Remember when celeron said that sort of thing should be a long-term goal? |
22:38 |
MTDiscord |
<greenxenith> We just disregarding that entirely? |
22:39 |
MTDiscord |
<greenxenith> No discussion on how this should actually be merged together sanely? |
22:40 |
MTDiscord |
<rollerozxa> I've just mostly been importing stuff from the voxelmanip wiki today, which includes the markdown conversion of the minetest_docs contents |
22:41 |
MTDiscord |
<greenxenith> I can see that, but minetest_docs is (was?) a separate project |
22:41 |
MTDiscord |
<greenxenith> No one ever formally discussed whether they should actually be the same thing. I've only proposed it, no one on the docs team bothered to comment on the issue |
22:42 |
MTDiscord |
<wsor4035> . |
22:43 |
MTDiscord |
<wsor4035> Should != could. It's just an existing content being brought in. |
22:44 |
MTDiscord |
<wsor4035> Also side note, not sure why I couldn't reply to that on mobile |
22:44 |
MTDiscord |
<greenxenith> If you formally supported immediately merging the two projects you could say that. Here. Or on the issue. |
22:45 |
MTDiscord |
<greenxenith> Throwing away the process for the sake of moving as fast as you possibly can might not turn out how you want |
22:46 |
MTDiscord |
<wsor4035> I will admit I was slightly 🤨 on it being brought in today, but figured 🤷 since the stuff was effectively dead |
22:46 |
MTDiscord |
<wsor4035> Maybe s/dead/inactive af |
22:47 |
MTDiscord |
<mark.wiemer> What is the goal of minetest_docs? What is the goal of dev.luanti.org? What are the key differences, if any? |
22:47 |
MTDiscord |
<greenxenith> This is what the issue covered |
22:47 |
MTDiscord |
<mark.wiemer> Would you mind re-linking the issue for clarity? |
22:47 |
MTDiscord |
<greenxenith> https://github.com/minetest/dev.luanti.org/issues/19 |
22:49 |
MTDiscord |
<rollerozxa> it's also much higher quality than most of the other stuff we brought in from the original developer wiki, and I think it sees a lot of value from being available in this format compared to being effectively hidden in the minetest_docs repo or on the voxelmanip wiki, which was intended to be a temporary storage until something like this very project happened |
22:50 |
MTDiscord |
<rollerozxa> minetest_docs was supposed to be a place where we could write supplementary documentation for the Lua API in a separate repository from the engine |
22:50 |
MTDiscord |
<greenxenith> ... no, it was intended to replace the Lua API docs |
22:50 |
MTDiscord |
<rollerozxa> well that was the original plan |
22:50 |
MTDiscord |
<greenxenith> It still is |
22:50 |
MTDiscord |
<greenxenith> That never changed |
22:53 |
MTDiscord |
<wsor4035> I don't think debating where the train was supposed to end is much good when it was dead on the tracks for a year? Plus |
22:53 |
MTDiscord |
<greenxenith> Well that was the question |
22:53 |
MTDiscord |
<greenxenith> Mark asked what the goal was |
22:55 |
MTDiscord |
<wsor4035> I think your issue is good for a long term goal if what to further grow to. The only that really happened was: * getting the docs out of a garbage format * making them accessible from a website * progress rather than bikesheeding like this |
22:56 |
MTDiscord |
<greenxenith> I despise the fact that this community has bikeshed so much we dont know what bikeshedding is any more |
22:56 |
MTDiscord |
<mark.wiemer> Minetest docs original goal: Replace the Lua API docs  Something from minetest_docs, in my book: going beyond API reference material and into guides/tutorials  Posted a big comment on https://github.com/minetest/dev.luanti.org/issues/19 |
22:58 |
MTDiscord |
<greenxenith> minetest_docs was already in a not-garbage format (AsciiDoc), simply being hosted on a third-party wiki (as MD) because they wanted to. That's fine. But all discussion on merging that content with this one was skipped. |
22:58 |
MTDiscord |
<greenxenith> And I say "not-garbage" to AsciiDoc in the sense of "it was fine where it was" not "there arent better things" |
22:59 |
MTDiscord |
<mark.wiemer> I'm new to projects like this, where we all contribute in our free time, and I think it's really easy to get lost in the weeds of what is and isn't bikeshedding. I see that having an endless thread of chat messages can lead to an endless circular discussion. My goal is to "set agendas", as boring as that sounds, so conversations can stay focused and folks can feel like we're all moving forward toward a common goal  That said I |
22:59 |
MTDiscord |
think if folks don't want to participate in discussions like "what's the goal of this project" they, well, don't really have to. The problem is when people start committing code that confuses others, that can ruin momentum and ultimately kill projects if folks aren't on the same page  Maybe I'm getting too meta, just my two cents |
23:00 |
MTDiscord |
<greenxenith> That is spot on. And exactly what is happening in other areas than just here (trying to move the project forward with zero direction or planning) |
23:01 |
MTDiscord |
<greenxenith> And I hate to be the guy pulling back on what feels like crazy good progress compared to what we are used to |
23:02 |
MTDiscord |
<greenxenith> I am one of the first people that will complain about the lack of direction or movement in this project as a whole, but I also believe in procedure and process to aid movement. |
23:03 |
MTDiscord |
<mark.wiemer> Yeah it's a hard spot. I think at this point we should just ask: are we OK making dev.luanti.org the "new standard" and working to merge all content from VoxelManip, minetest_docs, the Minetest wiki, and lua_api.md into it? If not, why not? Â I know it feels like the process was skipped in this case, but from what I've seen folks are all on the same page: dev.luanti.org is "the new standard", but maybe we've only heard from some |
23:03 |
MTDiscord |
folks in the past few days instead of everyone who has thoughts on it |
23:03 |
MTDiscord |
<mark.wiemer> (Sorry if I'm speaking with too much confidence. Obv I'm super new to this project and community. Please let me know if I'm "out of line" or off-base, just speaking from work experience as a software engineer, not sure how applicable it is outside of "work-work") |
23:04 |
MTDiscord |
<wsor4035> It might feel like something was skipped. But in reality talking about merging the wiki has happened for years, some behind closed doors, other in public, etc. It's more some people willing to do the work have shown up |
23:05 |
MTDiscord |
<greenxenith> I am not talking about the wiki, wsor |
23:05 |
MTDiscord |
<greenxenith> I am talking about minetest_docs |
23:05 |
MTDiscord |
<wsor4035> Yeah, and I'm responding to mark, not you |
23:06 |
MTDiscord |
<mark.wiemer> I think in this case there has been a lot of discussion but maybe no clear consensus in writing before things just "started happening", esp with dev.luanti.org? For me I have no expectations, being new, and I don't mean to speak on Xenith's behalf, but I can kinda see where he's coming from. Â Gotta walk the dog now, back in ~30 minutes |
23:06 |
MTDiscord |
<mark.wiemer> For what it's worth, I'm very OK making dev.luanti.org the "new standard" |
23:08 |
MTDiscord |
<wsor4035> Looking at worst case right? Say some agreement is reached on something that isn't dev.luanti.org. it can just be redirected to |
23:08 |
MTDiscord |
<greenxenith> Things "just happening" in order to get started is fine, and kind of how it has to happen around here. I am just concerned that the starting mentality may be carrying too strong into regular continuation |
23:09 |
MTDiscord |
<greenxenith> And once again, I am obviously a huge proponent of merging them all. I was the one who opened that issue in the first place. |
23:10 |
MTDiscord |
<greenxenith> Its just setting off a lot of red flags when we have active discussion issues that get bypassed almost immediately (the Minetest Wiki discussion, minetest_docs discussion) |
23:11 |
MTDiscord |
<wsor4035> Again, I was kinda 🤨 at first, but then decided I didn't care so much since it wasn't really going to hurt anything |
23:12 |
MTDiscord |
<greenxenith> I also saw the discussion earlier about removing PR requirements for trusted contributors, which would be fine after a project has found its footing and direction, but I am not certain this one has yet (clearly) |
23:12 |
MTDiscord |
<josiah_wi> I once posted a question on a pull request to an open source project asking about what I thought looked like a bug. My question was ignored and the pull request was merged. Then they discovered the bug I'd been asking about and had to go back and patch it. |
23:12 |
MTDiscord |
<wsor4035> This repo never had rules |
23:12 |
MTDiscord |
<josiah_wi> Leave other people's discussions unended and that is what will begin to happen. |
23:13 |
MTDiscord |
<greenxenith> Ok, it should. |
23:14 |
MTDiscord |
<josiah_wi> Can we at least give people a chance to object to merging the repositories? If no one objects I'd say merge them. |
23:14 |
MTDiscord |
<greenxenith> That was already skipped |
23:14 |
MTDiscord |
<wsor4035> I'm going to dm you green |
23:14 |
MTDiscord |
<greenxenith> They have been merged |
23:15 |
MTDiscord |
<josiah_wi> Ok, not much use arguing about it, then. |
23:15 |
MTDiscord |
<greenxenith> We arent |
23:15 |
MTDiscord |
<josiah_wi> But for future decisions let's not make big organizational changes without giving people the option to object to it. |
23:15 |
MTDiscord |
<greenxenith> That is what I am raising concerns about. The lack of procedure. |
23:16 |
MTDiscord |
<greenxenith> (For the third time, I am the one who suggested they be merged in the first place, so I obviously dont mind them being merged) |
23:16 |
MTDiscord |
<josiah_wi> Do we have procedures celeron55 and the core devs have in place that we can reflect? Green is right, we need procedure in place. It's not just about effective organization, it's about the health of the community. |
23:17 |
MTDiscord |
<josiah_wi> Worst case, someday someone will do something that makes somebody angry, whether it's worth being angry over or not, and it'll destroy the project and it'll die on the tracks with everyone refusing to contribute. |
23:17 |
MTDiscord |
<greenxenith> They dont have many procedures, and what they do have isnt particularly relevant to a documentation project |
23:19 |
MTDiscord |
<josiah_wi> Is it helpful for me to propose having celeron55 appoint someone to make organizational decisions, or present rules for how to appoint someone? |
23:20 |
MTDiscord |
<josiah_wi> Or two people. It just needs to be few enough people that if we start bike-shedding someone can authoritatively stop it and make a decision. |
23:22 |
MTDiscord |
<josiah_wi> On an unrelated note, can we put compilation docs in the new repo? I think it'd be an appropriate place to put the stuff Mark wrote, as well as additional information, potentially. |
23:35 |
MTDiscord |
<mark.wiemer> I think discussions in code repos are critical as things might break, but doc repos should be a bit more lax as things are probably 10x simpler compared to code. Â Having an authoritative "stop talking" guy sounds like that guy will become hated and not valuable, honestly. I'm more in favor of individual patience and maybe coming up with community guidelines instead of centralizing authority in some guy. BDFNs are great for some |
23:35 |
MTDiscord |
things but not for this, IMO. Â I think dev.luanti.org needs a mission statement, and that statement should be "the new standard for all Luanti documentation" Â That said, I think it is premature to start deleting/actively redirecting other wikis and doc pages to dev.luanti.org (DLO), as DLO is too new for that. Â Time for grocery shopping for me |
23:43 |
MTDiscord |
<wsor4035> It's worth clarifying that dlo isn't new, it's just bringing in existing content and centralizing it with a bit of house keeping regarding what not to touch to save effort |
23:44 |
MTDiscord |
<josiah_wi> @rollerozxa Are you planning to work on enforcing the front matter layout with Hugo? And it may be a lot to ask, but are you willing to teach me how that works? |
23:46 |
MTDiscord |
<greenxenith> wsor got this plane off the ground, so for right now he is the pilot until we reach cruising altitude. To be specific, once the initial plan is complete (getting the dev wiki and player/VM wiki merged), we can focus more on organization and further direction. This project is entirely under the Docs Team, so its up to us to communicate with each other and make compromises as we can. The initial takeoff might be rough, but that's |
23:46 |
MTDiscord |
fine, and what's most important is setting clear expectations. Whether or not we have contributing guidelines or merge rules right now, clear expectations will make this smoother. And right now the expectation is to keep momentum going through the initial hurdles that were planned. |
23:48 |
MTDiscord |
<josiah_wi> Ok, then in line with that objective, can we forgive wsor's merge? |
23:49 |
rubenwardy |
for your own sanity, I'd suggest limiting the scope and avoiding too much redundant information there that will quickly become outdated |
23:49 |
MTDiscord |
<greenxenith> IRC jumpscare |
23:52 |
MTDiscord |
<greenxenith> I think it was Roller's but yes, I am cool with it. I am sorry if it felt like I was criticizing @rollerozxa specifically for doing this or wsor for letting it happen. I don't blame either of them; my main concern was establishing procedures, but I didn't define that expectation clearly. Now that I understand others' expectations, I am more comfortable. |
23:54 |
MTDiscord |
<greenxenith> I believe that was an existing page on the original wiki? Im... actually not sure why it isnt here |
23:54 |
MTDiscord |
<greenxenith> It was definitely there before |
23:54 |
MTDiscord |
<greenxenith> https://cdn.discordapp.com/attachments/926231483155378176/1318003349836271676/image.png?ex=6760bdc5&is=675f6c45&hm=a9f151fc0a796294644cf6f7579cc83ede1a5a732bdea462f6de445e1312cae6& |
23:55 |
MTDiscord |
<josiah_wi> Wait, why do we have instructions on the Wiki and in the Luanti repo? |
23:56 |
MTDiscord |
<greenxenith> The wiki page may have pointed to the repo |
23:56 |
MTDiscord |
<greenxenith> This is one of those things that should probably live in the repo but be imported automatically into the wiki |
23:56 |
MTDiscord |
<josiah_wi> Oh, great, we do have instructions for MSYS2 on the new wiki. |
23:57 |
MTDiscord |
<greenxenith> Arguably that should be in https://github.com/minetest/minetest/blob/master/doc/compiling/windows.md but we can attend to that later |
23:58 |
MTDiscord |
<josiah_wi> https://dev.luanti.org/Compiling_Luanti came up when I did a Google search. |
23:58 |
MTDiscord |
<greenxenith> Yes, as seen above |
23:59 |
MTDiscord |
<greenxenith> It 404s because it doesnt exist on the migrated version. But that may be because the original only linked to the repo |
23:59 |
MTDiscord |
<greenxenith> rubenwardy: Is there information you are referring to in particular? |