Glider synthesis on LifeWiki

[GUYTU6J]

Here are some ambitious plans concerning how individual glider syntheses are stored and introduced on LifeWiki.
1)If you look at Special:WantedPages, you will see that many items after #733 are of the form RLE:<pname> synth. It means there are a lot of wanted synthesis RLE pages, regardless of whether or not a corresponding pattern have already been uploaded to conwaylife.com/patterns server. To clean up these red links, we need to create pages in bulk...
... but what pattern to choose? Mark Niemiec's database is outdated while Catagolue storage shows only one optimized result.
2)Thanks to muzik/AwesoMan3000, we now have a boilerplate template for general uninformative description of individual syntheses, putting every page transcluding it into a tracking category that asks for further information. It received complaint since inception and I talked that in the category talk page, but it seems to be stuck on LifeWiki demanding sizable amount of attention. We need to invite Mark Niemiec and Heinrich Koenig to take some time passing their knowledge to these articles including

• The discoverer and date of progressively reduced and other important glider syntheses,
• The relevant starting materials and components, and
• Special-purpose syntheses, especially good-clearance edge-shooters.

Are they feasible?

[dvgrn]

Here are some ambitious plans concerning how individual glider syntheses are stored and introduced on LifeWiki.
1)If you look at Special:WantedPages, you will see that many items after #733 are of the form RLE:<pname> synth. It means there are a lot of wanted synthesis RLE pages, regardless of whether or not a corresponding pattern have already been uploaded to conwaylife.com/patterns server. To clean up these red links, we need to create pages in bulk...

I can write something that does bulk page creation, once we figure out exactly what should be on the pages.

... but what pattern to choose? Mark Niemiec's database is outdated while Catagolue storage shows only one optimized result.

Some year relatively soon, we'll get a (public) update to Mark's database. In the meantime, I'd say that even an outdated version of one of Mark's stamp collections is very often hugely superior to a copy of the current best single synthesis from Catagolue. It's really useful to have lots of different ways to build the same thing, with different clearances and construction speeds and so on, all collected in one place.

So I'd suggest that if we really want to de-redlink the RLE:<pname> synth pages, we should collect all of the current relevant pages from Mark Niemiec's database and upload those -- or else continue to wait paitently for the next update to appear and do the work then, so that we don't have to re-do everything. Or build something that can be run automatically after the update, I suppose, and semi-automatically bring everything up to date without anyone having to do page-by-page comparisons.

The other alternative is to rework the LifeWiki to admit that glider synthesis tracking has now moved to Catagolue, and just take out the whole collection of uploaded synthesis patterns. But I don't think I want to do that, because Catagolue is terrible at providing the information about alternate syntheses that makes Mark's database so useful.

[confocaloid]

At least for "relatively small" syntheses of "relatively small" objects, what about documenting the syntheses of some pattern directly on the dedicated page for that pattern - that is, without touching the RLE namespace at all?

For example, see the current version of this page. Here is another example.

If I read this 2021 post correctly, there should be nothing wrong with putting RLEs of small patterns directly into the embedded viewer template.

[Book]

At least for "relatively small" syntheses of "relatively small" objects, what about documenting the syntheses of some pattern directly on the dedicated page for that pattern - that is, without touching the RLE namespace at all?

For example, see the current version of this page. Here is another example.

If I read this 2021 post correctly, there should be nothing wrong with putting RLEs of small patterns directly into the embedded viewer template.

There was a push last year to get rle's replaced by pnames so patterns would be in the wiki collection and there's a maint tracking report for that. However, I agree that smallish syntheses (especially as more and more are being shown in viewers) and other "illustrative" rles may not need to be memorialized in the collection. Seems like a judgment call.

[confocaloid]

There was a push last year to get rle's replaced by pnames so patterns would be in the wiki collection ...

Can you please provide links to relevant discussions?

[Book]

Can you please provide links to relevant discussions?

The discussion was sort of scattered around, but here's one post I found: viewtopic.php?f=3&t=5443&p=136934&hilit ... er#p136934

[confocaloid]

Crosspost

Is there a reason to still show the 8G synthesis? Why is it significant now that there's a 6G synthesis available?

When I earlier added the 8G synthesis, the intent was to create a place where subsequent improvements could be added, showing history of reductions. It is interesting to see how cheaper and cheaper glider syntheses were found over time. I think the old synthesis should be kept in the article, along with the link to Mark Niemiec's pages (which are a common source for many other glider syntheses for other targets as well).

A reader does not have to dig to find the syntheses. The footnote already provides two separate links -- one link to the overview page "Life Thirteen-Bit Still-Lifes" (which is necessary to give context), and another link "(download pattern file: 13/13-42.rle)" which leads directly to the synth file.

[confocaloid]

When earlier I edited a number of wiki pages about small still lives, I attempted to make the presentation more consistent, in particular by ensuring that the orientation of every still life shown in the infobox exactly matches the orientation of that still life in Mark Niemiec's synthesis database.

One of several recent undiscussed editing campaigns reverted several of those earlier changes, without explaining why.

I think the orientations of RLEs and images should be restored back, to agree with Niemiec's database.

[...]

Your recent changes are an example of "major changes affecting a large proportion of wiki content", which should be discussed first before trying to implement them.

Which changes exactly? Aside from the isotropic non-totalistic page split and the still life symmetries dispute (reply forthcoming), nothing from the past weeks appears to be controversial.

[...]
Unexplained reverts of an earlier attempt to match the orientation of the shown pattern that is given in Niemiec's database, in multiple RLE files including:

https://conwaylife.com/w/index.php?titl ... did=127465
https://conwaylife.com/w/index.php?titl ... did=127463
https://conwaylife.com/w/index.php?titl ... did=131284
https://conwaylife.com/w/index.php?titl ... did=116816
https://conwaylife.com/w/index.php?titl ... did=130638
https://conwaylife.com/w/index.php?titl ... did=142625
https://conwaylife.com/w/index.php?titl ... did=142626
https://conwaylife.com/w/index.php?titl ... did=127010
[...]

[muzik]

When earlier I edited a number of wiki pages about small still lives, I attempted to make the presentation more consistent, in particular by ensuring that the orientation of every still life shown in the infobox exactly matches the orientation of that still life in Mark Niemiec's synthesis database.

Was a consensus reached before a change this major was made?

One of several recent undiscussed editing campaigns reverted several of those earlier changes, without explaining why.

The reasoning here should be obvious: the orientations in RLEs were inconsistent with the orientations used in images, and this was an attempt to remedy said issue.

[...]
Unexplained reverts of an earlier attempt to match the orientation of the shown pattern that is given in Niemiec's database, in multiple RLE files including:

https://conwaylife.com/w/index.php?titl ... did=127465
https://conwaylife.com/w/index.php?titl ... did=127463
https://conwaylife.com/w/index.php?titl ... did=131284
https://conwaylife.com/w/index.php?titl ... did=116816
https://conwaylife.com/w/index.php?titl ... did=130638
https://conwaylife.com/w/index.php?titl ... did=142625
https://conwaylife.com/w/index.php?titl ... did=142626
https://conwaylife.com/w/index.php?titl ... did=127010
[...]

I can't help but notice that in this list, you've specifically picked all of the edits that happened to show "Manual revert", and not the roughly 100 others that didn't coincidentally end up undoing a prior change.

[confocaloid]

When earlier I edited a number of wiki pages about small still lives, I attempted to make the presentation more consistent, in particular by ensuring that the orientation of every still life shown in the infobox exactly matches the orientation of that still life in Mark Niemiec's synthesis database.

Was a consensus reached before a change this major was made?
[...]

Unlike your recent editing campaign, those changes were not major, and affected only several pages.

As far as I remember, before that, the majority of pages did already use orientation that matched Niemiec's database.

As far as I know, nobody objected to those changes at the time.

[...]

One of several recent undiscussed editing campaigns reverted several of those earlier changes, without explaining why.

The reasoning here should be obvious: the orientations in RLEs were inconsistent with the orientations used in images, and this was an attempt to remedy said issue.
[...]

When the orientation of an object in an RLE does not match the orientation of the object in the corresponding image file, one simply has to check which orientation is the preferred one, and change either the RLE, or the image, or both.

When the RLE already uses the same orientation as Niemiec's database, and the image file uses some other orientation, that means one needs to change the image.

[...]

[...]
Unexplained reverts of an earlier attempt to match the orientation of the shown pattern that is given in Niemiec's database, in multiple RLE files including:

https://conwaylife.com/w/index.php?titl ... did=127465
https://conwaylife.com/w/index.php?titl ... did=127463
https://conwaylife.com/w/index.php?titl ... did=131284
https://conwaylife.com/w/index.php?titl ... did=116816
https://conwaylife.com/w/index.php?titl ... did=130638
https://conwaylife.com/w/index.php?titl ... did=142625
https://conwaylife.com/w/index.php?titl ... did=142626
https://conwaylife.com/w/index.php?titl ... did=127010
[...]

I can't help but notice that in this list, you've specifically picked all of the edits that happened to show "Manual revert", and not the roughly 100 others that didn't coincidentally end up undoing a prior change.

In other words, you undertook a large-scale editing campaign, apparently without bothering to check whether there is a consensus for editing those "roughly 100 others" pages in the specific way you decided to do.

Apparently you didn't bother to look to the edit history of those RLEs to see whether there were earlier changes in the other direction, either. Any such earlier changes in the opposite direction are an additional reason to avoid editing anything, and instead open a discussion regarding the issue. Which you didn't do.

[confocaloid]

The wiki edit https://conwaylife.com/w/index.php?diff ... did=152738 involves an undiscussed and unexplained removal of the "after" parts from synthesis RLEs.
The same applies to wiki edit https://conwaylife.com/w/index.php?diff ... did=154784

When I added syntheses, I tried to follow Niemiec's database, by showing both the synthesis and the resulting object.

[galoomba]

The wiki edit https://conwaylife.com/w/index.php?diff ... did=152738 involves an undiscussed and unexplained removal of the "after" parts from synthesis RLEs.
The same applies to wiki edit https://conwaylife.com/w/index.php?diff ... did=154784

When I added syntheses, I tried to follow Niemiec's database, by showing both the synthesis and the resulting object.

Well, the wiki is not Niemiec's database. In particular, the syntheses on the wiki aren't just files in a database; they're embedded on the page for that particular pattern, so it's already clear what pattern they make. They're also animated, so the "after" does in fact appear during the course of the animation. The second copy of the object is redundant and clutters the animation.

As for "undiscussed and unexplained", I don't know why you think that people should have to ask you for your personal opinion before making any kind of wiki edit. We've had this discussion many times already.

[hotdogPi]

As for "undiscussed and unexplained", I don't know why you think that people should have to ask you for your personal opinion before making any kind of wiki edit. We've had this discussion many times already.

There's no need to ask for approval, but if there's an objection, it should be reverted and sorted out. See WP:BOLD.

Neither user did anything wrong here.

I don't have an opinion either way on which way is preferable regarding the content of the debate.

[confocaloid]

[...] Well, the wiki is not Niemiec's database. In particular, the syntheses on the wiki aren't just files in a database; they're embedded on the page for that particular pattern, so it's already clear what pattern they make. They're also animated, so the "after" does in fact appear during the course of the animation. The second copy of the object is redundant and clutters the animation. [...]

The main usefulness of the "after" part of glider syntheses isn't in making it "clear what pattern they make".
The "after" part helps to line up different stages, when making an incremental synthesis. This is a big part of what makes it helpful.

I disagree with your claim that the "after" part clutters the animation. I think the "after" part is useful and should be preserved in the wiki as well.

[...] As for "undiscussed and unexplained", I don't know why you think that people should have to ask you for your personal opinion before making any kind of wiki edit. [...]

Again: your recent edits are an undiscussed and unexplained large-scale editing campaign, and moreover an editing campaign that undoes changes previously made by someone else. Specifically, you are undoing my earlier efforts to add the "after" part to glider syntheses shown in the wiki.

It's obvious that large-scale "undo" changes of this kind should be discussed and agreed upon, before they are made.

[galoomba]

[...] Well, the wiki is not Niemiec's database. In particular, the syntheses on the wiki aren't just files in a database; they're embedded on the page for that particular pattern, so it's already clear what pattern they make. They're also animated, so the "after" does in fact appear during the course of the animation. The second copy of the object is redundant and clutters the animation. [...]

The main usefulness of the "after" part of glider syntheses isn't in making it "clear what pattern they make".
The "after" part helps to line up different stages, when making an incremental synthesis. This is a big part of what makes it helpful.

I hadn't thought of that. Although not all of the ones I removed were lined up. Perhaps a way to preserve this while also reducing clutter would be to put them a bit farther away and hide them out of frame?
Although that might be more trouble than it's worth. How many people actually use the wiki's embedded patterns for incremental synthesis, and how much benefit does the "after" pattern provide? I can't see it being many/much, but maybe I'm wrong. If it saves someone a few seconds every few months, it doesn't really matter, and would definitely not be worth spending hours making the changes.

Again: your recent edits are an undiscussed and unexplained large-scale editing campaign, and moreover an editing campaign that undoes changes previously made by someone else. Specifically, you are undoing my earlier efforts to add the "after" part to glider syntheses shown in the wiki.

No, most of them weren't changes, they were added with the "after" part already included. Changing formatting is not the same thing as an undo.

[confocaloid]

[...] Perhaps a way to preserve this while also reducing clutter would be to put them a bit farther away and hide them out of frame? [...]

I disagree with your claim that the "after" part is "clutter".

I believe showing the "after" part is useful and not clutter.
It's useful both because of helping people who copy the syntheses from the wiki, and also because it's helpful to see the target object nearby (so that when the collision settles, there are two identical objects shown side-by-side).

[...] If it saves someone a few seconds every few months, it doesn't really matter, and would definitely not be worth spending hours making the changes.

If it saves some readers a few seconds every few months, then it is useful to some readers.
The "after" part is useful, there is no need to remove it.
(And no, it does not take "hours" to add it. IIRC I did it with about five minutes per synthesis, including making related changes and previewing the page.)

As a side note, your own logic apparently suggests that it's not worth "spending hours making the changes", making it unclear why you decided to do exactly that (i.e. make the changes to remove the target object that was already there, added by someone other than you).

[...]

Again: your recent edits are an undiscussed and unexplained large-scale editing campaign, and moreover an editing campaign that undoes changes previously made by someone else. Specifically, you are undoing my earlier efforts to add the "after" part to glider syntheses shown in the wiki.

No, most of them weren't changes, they were added with the "after" part already included. Changing formatting is not the same thing as an undo.

You are undoing my earlier efforts to add the "after" part to glider syntheses.
Clearly there was no way for me to add the "after" part accidentally. I wanted to add the "after" part, because I believe that it is helpful.

(Compare and contrast with minor formatting changes, where people could actually do something "accidentally" without thinking about it.)

[galoomba]

I disagree with your claim that the "after" part is "clutter".

I believe showing the "after" part is useful and not clutter.
It's useful both because of helping people who copy the syntheses from the wiki, and also because it's helpful to see the target object nearby (so that when the collision settles, there are two identical objects shown side-by-side).

I don't see the point, but i guess that's just a difference of opinion. Would be nice to get the opinion of a few more people here then.

(And no, it does not take "hours" to add it. IIRC I did it with about five minutes per synthesis, including making related changes and previewing the page.)

Five minutes per synthesis times the number of syntheses on the wiki adds up to hours. That's what I meant.

As a side note, your own logic apparently suggests that it's not worth "spending hours making the changes", making it unclear why you decided to do exactly that (i.e. make the changes to remove the target object that was already there, added by someone other than you).

That's because I'm making changes to make the pages cleaner, not to save some hypothetical incremental synthesis maker a few seconds.

You are undoing my earlier efforts to add the "after" part to glider syntheses.
Clearly there was no way for me to add the "after" part accidentally. I wanted to add the "after" part, because I believe that it is helpful.

For all I knew you could've just copied the syntheses from Niemiec. After all, your initial reason for including the "after" parts was that it's how Niemiec formats it, not that it's helpful.

[confocaloid]

[...] Would be nice to get the opinion of a few more people here then. [...]

Agreed.

[...] That's because I'm making changes to make the pages cleaner, [...]

As an example, here is a link to one of multiple edits you made recently: https://conwaylife.com/w/index.php?diff ... did=155118

This change doesn't look to me like "making the page cleaner".
Here is the revision before your edit: https://conwaylife.com/w/index.php?oldid=155118
Here is the revision after your edit: https://conwaylife.com/w/index.php?oldid=155946

The revision before your edit had a 4G synthesis and two 5G syntheses formatted in a consistent way, all showing the target object, all constructing the target object in the same orientation.

You changed that by discarding the "after" parts (so it becomes hard to see the object that's constructed), arbitrarily changing the orientations of syntheses, and trimming previously existing LifeViewer animation configuration. Neither of these changes feels like "making the page cleaner" to me.

For some reason, there is no edit summary for your linked edit. (Therefore, "unexplained".)

I think the same can be said about other your recent edits in pages showing glider syntheses: https://conwaylife.com/w/index.php?targ ... tributions

Neither of those changes was previously discussed. (Therefore, "undiscussed".)

[galoomba]

[...] That's because I'm making changes to make the pages cleaner, [...]

As an example, here is a link to one of multiple edits you made recently: https://conwaylife.com/w/index.php?diff ... did=155118

This change doesn't look to me like "making the page cleaner".

Ok, you're going a tangent based on the specific wording I used in a reply to what was already a tangent. But I suppose it's relevant, so I'll explain it anyway.

(so it becomes hard to see the object that's constructed)

That's a ridiculous statement. The object is clearly visible for several seconds after the synthesis finishes, and if that's not enough, you have a nice big picture of the object in the infobox on the same page.

arbitrarily changing the orientations of syntheses,

I changed the orientation of some to fit in the space better, and to be honest, I like seeing the object in different orientations. It feels like you're just looking for things to complain about at this point.

and trimming previously existing LifeViewer animation configuration.

The constant pauses were jarring and disorienting. I'm not the only person who has issues with them, as evidenced by the other time we discussed this.

[confocaloid]

[...] The object is clearly visible for several seconds after the synthesis finishes, and if that's not enough, you have a nice big picture of the object in the infobox on the same page. [...]

Neither of that by itself suffices to make verification of the synthesis as simple as possible.

In the revision before your changes ( https://conwaylife.com/w/index.php?oldid=155118 ), it's easy to see that the constructed object indeed coincides with the "after" part, precisely because both are shown side-by-side in the same orientation.
In the same revision, it is also easy to see that the "after" part coincides with the object that is the topic of the article, precisely because the "after" part does not change during animation, and it is straightforward to compare it against other images of the object in the article.

In other words, the revision before your changes makes it easy to see that the synthesis actually constructs the object that's the topic of the article, in two visual comparisons, each of which is easy to do.

In the revision after your changes ( https://conwaylife.com/w/index.php?oldid=155946 ), the "after" part is removed. That means that it is no longer possible to perform the quick comparison between the actually constructed object on the left and the expected object shown on the right, in the timespan before the animation resets.
Also, it is no longer possible to perform the other comparison, between the static "after" part and another image of the same object in the page (e.g. the infobox).

The bottom line is that, considering the effects of animation, your changes make it significantly harder for readers to see that the synthesis indeed works and constructs the correct object.
I believe your removals of the "after" part in those pages are significant damage to the wiki.

[...] I changed the orientation of some to fit in the space better, and to be honest, I like seeing the object in different orientations. [...]

Inconsistent orientations of the same object in the same page are very confusing. It becomes hard to compare different images of the same object. It becomes impossible to understand why these specific orientations were chosen (because apparently it ends up being a personal preference of some editor, who didn't even bother to explain their changes in the edit summaries, so that the community can understand what's going on, and didn't bother to discuss large-scale changes before making them).

[dvgrn]

Okay, looks like the issue at hand is fifteen removals of sample objects in syntheses, on these article pages:

Trans-block_on_long_bookend
Trans-loaf_with_tail
Block_on_cap
Elevener
Tapeworm
Trans-long_boat_with_tail
Beehive_with_hooked_tail
Cis-boat_on_cap
Long_boat_tie_ship
Beehive_on_table
Snorkel_loop
Claw_with_tail
Cis-loaf_with_tail
Cis-loaf_on_table
Cis-block_on_long_bookend

We're in step 3 of LW:DR dispute resolution rules at the moment. LifeWiki editing has a bias toward action when there are no objections, but more of a bias toward the status quo when there are disagreements.

So I'm going ahead and following hotdogPi's suggestion from yesterday -- that's the part of step 3 about "The content may be reverted to its pre-dispute state for the duration of the dispute.". That's not a judgment for or against the changes that galoomba made -- they were all perfectly fine good-faith edits at the time they were made. The rollback is just an acknowledgment that it makes sense to hold off on making those changes, until and unless a community consensus can be demonstrated in favor of them.

I might have missed some unrelated edits that will get rolled back along with these fifteen sets of edits. However, everything I saw in a quick survey was related to the removal of the sample objects. E.g., the time before looping the animation was often extended, so that the constructed object would be visible for a good chunk of time. That adjustment is not needed so much when the sample object is visible for the whole animation cycle.

[confocaloid]

The issue at hand is larger than just removals of "after" parts (sample objects) in glider syntheses.

Okay, looks like the issue at hand is fifteen removals of sample objects in syntheses, on these article pages:
[...]
We're in step 3 of LW:DR dispute resolution rules at the moment. LifeWiki editing has a bias toward action when there are no objections, but more of a bias toward the status quo when there are disagreements.

So I'm going ahead and following hotdogPi's suggestion from yesterday -- that's the part of step 3 about "The content may be reverted to its pre-dispute state for the duration of the dispute.". [...]
[...]

As I posted earlier in the forum thread "Glider synthesis on LifeWiki", one of several recent editing campaigns is a re-orientation of multiple RLEs in the "RLE:" namespace, without any explanations of the changes in the edit summaries, and without any preceding discussion.

I think those undiscussed and unexplained re-orientations of RLEs also should be reverted:

• reorientations of RLEs by User:Muzik: https://conwaylife.com/w/index.php?targ ... tributions
• reorientations of RLEs by User:Galoomba: https://conwaylife.com/w/index.php?targ ... tributions

As I wrote, when earlier I edited a number of wiki pages about small still lives, I attempted to make the presentation more consistent; in particular I attempted to ensure that the orientation of an object shown in the infobox matches the orientation of that object in syntheses included in Mark Niemiec's synthesis database.

Recent re-orientations of RLEs had the effect of undoing part of those my earlier efforts to make the presentation more consistent, while failing to provide any evidence of consensus for changing the orientation of objects, and also failing to provide any explanation for the new choices.

This is directly related to the presentation of glider syntheses in the affected pages, since it is clearly helpful to ensure that the same orientation of an object is used consistently through the entire page, including the infobox and the embedded viewers.

[...]

[...] I changed the orientation of some to fit in the space better, and to be honest, I like seeing the object in different orientations. [...]

Inconsistent orientations of the same object in the same page are very confusing. It becomes hard to compare different images of the same object. It becomes impossible to understand why these specific orientations were chosen (because apparently it ends up being a personal preference of some editor, who didn't even bother to explain their changes in the edit summaries, so that the community can understand what's going on, and didn't bother to discuss large-scale changes before making them).

[dvgrn]

The issue at hand is larger than just removals of "after" parts (sample objects) in glider syntheses.
...
As I posted earlier in the forum thread "Glider synthesis on LifeWiki", one of several recent editing campaigns is a re-orientation of multiple RLEs in the "RLE:" namespace, without any explanations of the changes in the edit summaries, and without any preceding discussion.

Yup -- really the issue is a lot larger than that, too. There are quite a few interlocking pieces here.

Along with the questions that have come up in the past couple of days --

"should sample objects be included in synthesis animation patterns?"

and now

"should the orientation of constructed objects in Mark Niemiec's database be used as the standard orientation for all representations of those objects on the LifeWiki -- images, infobox RLE, and synthesis RLE?"

-- there are also several other standardization questions that have been causing this same kind of back-and-forth editing according to different people's opinions, because there's never been any general agreement on what the standard is:

"should the orientation of the synthesized object match the orientation shown in the infobox?"

"should the bounding box of syntheses be kept as small as possible (for visibility), or is it more important to leave a clear wide separation between synthesis and sample object (to avoid confusion)?"

"should BOOK theme be used for synthesis animations, or should we stick with the old standard LifeWiki two-state patterns?"

A serious competitor for an orientation standard
Mark Niemiec's database used to be pretty much the only reasonable source for a standard orientation of objects -- but not any more! We already have Catagolue links for every object in LifeWiki articles. The orientation represented by the apgcode of an object seems like a really good alternate candidate for a "standard" orientation.

Time for a pause to discuss
The case for reverting RLE orientation changes is a lot more complicated than the case for rolling back galoomba's well-intentioned sample object removals. I didn't roll back muzik's orientation changes when confocaloid originally made that complaint, and I would recommend that nobody try rolling any RLE orientation changes back right now. The reason for this is that after rollback, the LifeWiki wouldn't be in any more of a consistent state than it is already.

The problem is that confocaloid's original RLE orientation changes were in incomplete effort. They left a lot of technical debt, in the form of existing uploaded images that no longer matched the orientations shown in the RLE:

...the orientations in RLEs were inconsistent with the orientations used in images, and this was an attempt to remedy said issue.

As far as I know (it's a lot of work even to check this point!) confocaloid's work also didn't involve a complete survey of Niemiec's database and the LifeWiki. If there are some articles that were never updated in the first place, where the object doesn't match the Niemiec database orientation, then that's another inconsistency that would have to be resolved once the community adopts a standard.

At the moment, let's just acknowledge that there are quite a few interrelated issues here, and that this thread is a good place to discuss them. Before any more back-and-forth changes get made, it would be good to figure out if it's possible to develop some kind of consensus about what standard practice should be..

[confocaloid]

[...] We already have Catagolue links for every object in LifeWiki articles. The orientation represented by the apgcode of an object [...]

I strongly disagree. Following "apgcode orientations" feels like a particularly bad idea. These orientations look arbitrary from the viewpoint of a human reader.
Even if objects are visually similar / belong to the same family, their "apgcode orientations" can be different.

Look what happens with snakes, for example:

Code: Select all

x = 36, y = 8, rule = B3/S23
2o8b2o8b2o12b2o$o9bobo7bobo12bo$bo11bobo7bo10bo$2bo11b2o8bobo6bo$b2o
22b2o5bo$31bo$30bo$30b2o!

Mechanically applying these orientations on the wiki leads to failures to be consistent, where consistency is possible.

I believe Niemiec's synthesis files are a much better source for orientations of notable objects, compared to Catagolue.

[hotdogPi]

I generally try for width to be larger than height just because most viewers aren't square.