Options:
Previous day, Next day
- # Session Start: Fri Oct 16 00:00:00 2015
- # Session Ident: #css
- # [00:00] * Quits: dbaron (~dbaron@public.cloak) ("8403864 bytes have been tenured, next gc will be global.")
- # [00:00] <fantasai> http://software.hixie.ch/utilities/js/live-dom-viewer/?saved=3682
- # [00:00] <fantasai> The container is as wide as two boxes
- # [00:00] * Joins: dbaron (~dbaron@public.cloak)
- # [00:00] <fantasai> because the third box isn't allowed to move leftwards of the first box's right edge
- # [00:00] <fantasai> We do *not* want that rule in flexbox
- # [00:01] <fantasai> and that explains why the containing block floors at 20px -- it's following the same rule
- # [00:01] <fantasai> floats are not linear
- # [00:02] <fantasai> they're "every float for itself"
- # [00:03] <TabAtkins> I'm... okay with that rule in Flexbox?
- # [00:03] <TabAtkins> Like, block's behavior that the third just follows the second up into the stratosphere is bizarre.
- # [00:03] <fantasai> It totally makes sense, I have no idea why you have a problem with this
- # [00:03] <fantasai> You place one box after the other box
- # [00:03] <fantasai> you don't care about where earlier boxes are because your reference edge is the previous box
- # [00:04] <fantasai> It's very simple
- # [00:04] <fantasai> It makes sense with negative sizes
- # [00:04] <fantasai> and it is linear as opposed to quadratic
- # [00:04] <TabAtkins> That's a possible rule to follow, yes. Another possible rule to follow is that you do that, but box sizes are floored at zero for the purpose of the container's layout.
- # [00:04] <TabAtkins> Why would the latter be quadratic?
- # [00:04] <fantasai> *that does not make sense and is not what floats do anyway*
- # [00:05] <TabAtkins> Sure. It's *similar* to what they do. (I'd be interested in seeing an example that shows a difference.)
- # [00:06] <fantasai> Look at that example again. If you do box-follows-box and *also* do "floor boxes at zero before calculating container as its sum"
- # [00:06] <fantasai> then that example ends up being as wide as two boxes
- # [00:06] <fantasai> but the third box gets shifted 40px to the left
- # [00:06] <fantasai> so you have this 40px gap in the box
- # [00:06] <fantasai> *THAT DOESN"T MAKE SENSE*
- # [00:06] <TabAtkins> ...why would it get shifted 40px left?
- # [00:07] <fantasai> because the previous box was 40px to the left
- # [00:07] <fantasai> and you said we're doing box-follows-box
- # [00:07] <fantasai> so, the third box moves leftward 40px so its left edge lines up with the second boxe's right edge
- # [00:07] <TabAtkins> Yeah, obviously, *mixing* behaviors (boxes always immediately follow preceding boxes, *and* floor at zero) doesn't make sense.
- # [00:07] <TabAtkins> When did I say we we're doing box-follows-box?
- # [00:08] <fantasai> "Another possible rule to follow is that you do that, but box sizes are floored at zero for the purpose of the container's layout.
- # [00:08] <fantasai> "
- # [00:08] <TabAtkins> "for the purpose of the container's layout" includes positioning the container's children.
- # [00:08] <fantasai> I dereferenced "that" to the most immediately previously-mentioned behavior
- # [00:08] <TabAtkins> So it treats the second child as zero for the purpose of placing the third child.
- # [00:08] <fantasai> That's dumb.
- # [00:08] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [00:08] <TabAtkins> Okay?
- # [00:09] <fantasai> What's the point of negative margins if you're flooring them at zero
- # [00:09] <TabAtkins> It's, as far as I can tell, indistinguishable from teh float behavior?
- # [00:09] <TabAtkins> What's the point of negative margins so large they overwhelm the rest of the box's size in the first place?
- # [00:09] <fantasai> That's not your problem
- # [00:09] <fantasai> that's the author's problem
- # [00:09] <TabAtkins> Sure.
- # [00:09] <fantasai> and maybe in that particular design
- # [00:09] <fantasai> it's not a problem
- # [00:09] <TabAtkins> I doubt it's ever not a problem.
- # [00:09] <fantasai> It should behave like block layout
- # [00:09] <fantasai> not like floats
- # [00:10] <TabAtkins> You keep saying that, but you haven't actually established a reason *why* the block behavior is better than the float behavior.
- # [00:10] <TabAtkins> Why not allow floats the same behavior?
- # [00:10] <TabAtkins> Their rules happen to pop out something different, but why is that better for them?
- # [00:10] <fantasai> because it doesn't have an arbitrary limit
- # [00:10] <fantasai> for no good reason
- # [00:10] <TabAtkins> ?
- # [00:10] <TabAtkins> What arbitrary limit?
- # [00:10] <fantasai> Floats do something different because they're following a different rule
- # [00:10] <TabAtkins> Yes.
- # [00:11] <fantasai> Each box must be to one side of *every box before it"*
- # [00:11] <fantasai> That's reasonable for floats
- # [00:11] <fantasai> It's also quadratic
- # [00:11] <fantasai> and it's not a necessary behavior here
- # [00:11] <TabAtkins> No it's not. It's only quadratic in an excessively naive impl.
- # [00:11] <fantasai> Zero
- # [00:11] <fantasai> Okay, whatever
- # [00:11] <fantasai> I still don't agree with this
- # [00:11] <fantasai> and I don't think anything in the spec says that you should do this
- # [00:11] <TabAtkins> Same as block would be quadratic if the rule was "each box's top must be placed at the sum of the previous boxe's heights".
- # [00:11] <fantasai> and I am against changing the spec to say we should do this.
- # [00:12] * fantasai is done
- # [00:12] <TabAtkins> lol ok, but I'm taking it to the list
- # [00:12] <fantasai> You want it changed, file an issue and get the WG to agree.
- # [00:12] <TabAtkins> because I'm *not* certain the behavior is right.
- # [00:14] <fantasai> Anyway, let's work on something more useful than arguing over overly-negative boxes ?
- # [00:14] <fantasai> :)
- # [00:14] * fantasai doesn't have much time today
- # [00:14] <TabAtkins> *you're* an overly negative box.
- # [00:14] * Quits: liam (liam@public.cloak) (Ping timeout: 180 seconds)
- # [00:14] <TabAtkins> Yeah, let's see what else needs to be done...
- # [00:19] * Joins: shepazu (schepers@public.cloak)
- # [00:35] <adenilson> TabAtkins: ping? Thanks for merging the css-round-display preposition fix. If you think is fine, there is another one at: https://github.com/w3c/csswg-drafts/pull/59
- # [00:36] <adenilson> and should I submit a merge concerning the re-phrasing of the note? i.e. “If the shapes of displays are various, ...” -> “For other shapes, …”
- # [00:46] <TabAtkins> Yeah, will check the others, just happened across the first while doing something else. Busy right now. ^_^
- # [00:55] <adenilson> Coolio.
- # [01:26] * Quits: dbaron (~dbaron@public.cloak) ("8403864 bytes have been tenured, next gc will be global.")
- # [01:26] * Joins: dbaron (~dbaron@public.cloak)
- # [01:34] * Joins: Florian (~Florian@public.cloak)
- # [01:36] * Quits: jdaggett (~jdaggett@public.cloak) (jdaggett)
- # [01:42] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [01:51] * Quits: dauwhe (~dauwhe@public.cloak) (Client closed connection)
- # [01:56] * heycam is now known as heycam|away
- # [01:58] * Joins: Florian (~Florian@public.cloak)
- # [02:10] * Joins: tantek (~tantek@public.cloak)
- # [02:19] * Joins: dauwhe (~dauwhe@public.cloak)
- # [02:24] * Joins: liam (liam@public.cloak)
- # [02:26] * Quits: dauwhe (~dauwhe@public.cloak) (Ping timeout: 180 seconds)
- # [02:52] * Quits: adenilson (~anonymous@public.cloak) (adenilson)
- # [03:04] * Quits: darktears (~darktears@public.cloak) ("Leaving...")
- # [03:26] * Joins: jdaggett (~jdaggett@public.cloak)
- # [03:41] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [03:44] * Joins: dauwhe (~dauwhe@public.cloak)
- # [04:12] * Quits: dbaron (~dbaron@public.cloak) ("8403864 bytes have been tenured, next gc will be global.")
- # [05:31] * Quits: myles (~Adium@public.cloak) ("Leaving.")
- # [05:35] * Joins: tantek (~tantek@public.cloak)
- # [06:05] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [06:48] * Joins: tantek (~tantek@public.cloak)
- # [07:22] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [07:22] * Joins: Florian (~Florian@public.cloak)
- # [07:23] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [07:23] * Joins: Florian (~Florian@public.cloak)
- # [07:30] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [07:30] * Joins: Florian (~Florian@public.cloak)
- # [07:33] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [07:35] * Joins: tantek (~tantek@public.cloak)
- # [07:40] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [07:45] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [07:45] * Joins: Florian (~Florian@public.cloak)
- # [08:26] * Quits: jdaggett (~jdaggett@public.cloak) (jdaggett)
- # [08:51] * Joins: tantek (~tantek@public.cloak)
- # [09:20] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [09:31] * Joins: zcorpan (~zcorpan@public.cloak)
- # [10:00] * Joins: antonp (~Thunderbird@public.cloak)
- # [10:04] * Joins: Ms2ger (~Ms2ger@public.cloak)
- # [10:10] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [10:11] * Joins: lajava (~javi@public.cloak)
- # [10:21] * Joins: zcorpan (~zcorpan@public.cloak)
- # [10:23] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [10:32] * Joins: jdaggett (~jdaggett@public.cloak)
- # [11:35] * Joins: zcorpan (~zcorpan@public.cloak)
- # [11:35] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [11:35] * Joins: zcorpan (~zcorpan@public.cloak)
- # [12:01] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [12:02] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [12:03] * Quits: antonp (~Thunderbird@public.cloak) (Client closed connection)
- # [12:03] * Joins: antonp (~Thunderbird@public.cloak)
- # [12:04] * Joins: Florian (~Florian@public.cloak)
- # [12:04] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [12:04] * Joins: Florian (~Florian@public.cloak)
- # [12:19] * Quits: jdaggett (~jdaggett@public.cloak) (jdaggett)
- # [12:33] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [12:53] * Joins: zcorpan (~zcorpan@public.cloak)
- # [12:58] * Joins: Florian (~Florian@public.cloak)
- # [12:58] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [12:58] * Joins: Florian (~Florian@public.cloak)
- # [13:01] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [13:25] * Joins: zcorpan (~zcorpan@public.cloak)
- # [13:38] * Joins: jdaggett (~jdaggett@public.cloak)
- # [14:21] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [14:21] * Joins: Florian (~Florian@public.cloak)
- # [14:21] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [14:22] * Joins: Florian (~Florian@public.cloak)
- # [14:23] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [14:28] * Joins: plh (plehegar@public.cloak)
- # [14:52] * Quits: jdaggett (~jdaggett@public.cloak) (jdaggett)
- # [15:02] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [15:02] * Joins: plh (plehegar@public.cloak)
- # [15:06] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [15:24] * Joins: Florian (~Florian@public.cloak)
- # [15:25] * Joins: Florian_ (~Florian@public.cloak)
- # [15:25] * Quits: Florian_ (~Florian@public.cloak) (Client closed connection)
- # [15:25] * Joins: Florian_ (~Florian@public.cloak)
- # [15:28] * Joins: javifgb (~javi@public.cloak)
- # [15:28] * Quits: javifgb (~javi@public.cloak) (Client closed connection)
- # [15:31] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [15:54] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [15:54] * Joins: plh (plehegar@public.cloak)
- # [17:01] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [17:02] * Joins: plh (plehegar@public.cloak)
- # [17:20] * Joins: darktears (~darktears@public.cloak)
- # [17:47] * Quits: Ms2ger (~Ms2ger@public.cloak) ("nn")
- # [17:49] * Quits: darktears (~darktears@public.cloak) ("Linkinus - http://linkinus.com")
- # [17:49] * Joins: darktears (~darktears@public.cloak)
- # [18:12] * Quits: Florian_ (~Florian@public.cloak) (Client closed connection)
- # [18:12] * Joins: Florian (~Florian@public.cloak)
- # [18:13] * Quits: Florian (~Florian@public.cloak) (Client closed connection)
- # [18:13] * Joins: Florian (~Florian@public.cloak)
- # [18:20] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [18:47] * Joins: tantek (~tantek@public.cloak)
- # [19:13] * Joins: Florian (~Florian@public.cloak)
- # [19:16] * Quits: antonp (~Thunderbird@public.cloak) (antonp)
- # [19:20] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [19:43] * Joins: myles (~Adium@public.cloak)
- # [19:56] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [20:17] * Quits: gsnedders (~gsnedders@public.cloak) ("leaving")
- # [20:20] * Joins: gsnedders (~gsnedders@public.cloak)
- # [20:20] * Joins: plh (plehegar@public.cloak)
- # [20:22] * Joins: adenilson (~anonymous@public.cloak)
- # [20:23] <cbiesinger> TabAtkins: fantasai: fyi, current FF and blink impl do not interoperate on this negative margin issue
- # [20:24] <TabAtkins> Unsurprising! For Flexbox specifically?
- # [20:24] <cbiesinger> TabAtkins: yes, though I didn't test non-flexbox cases
- # [20:27] * Joins: zcorpan (~zcorpan@public.cloak)
- # [20:27] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [20:27] * Joins: zcorpan (~zcorpan@public.cloak)
- # [20:32] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [20:41] * Joins: zcorpan (~zcorpan@public.cloak)
- # [20:43] * Quits: tantek (~tantek@public.cloak) (tantek)
- # [20:44] <cbiesinger> TabAtkins: should I email www-style for the negative margin thing? it sounds like you and elika didn't agree
- # [20:45] <TabAtkins> I'm finishing up an email I started yesterday now.
- # [20:45] <cbiesinger> ah great
- # [20:45] <TabAtkins> If you could send me your testcase demonstrating diff behavior, that'd be great.
- # [20:45] <cbiesinger> thanks :)
- # [20:45] <cbiesinger> sure one sec, though the pull request should work too
- # [20:49] <cbiesinger> hmm maybe I was just getting confused
- # [20:49] <cbiesinger> that testcase is interoperable??
- # [20:51] <cbiesinger> TabAtkins: ok as far as I can tell I was wrong, chrome and firefox are currently interoperable
- # [20:51] <cbiesinger> http://plexode.com/eval3/#s=aekVQXANJVQMbAx1yAXgePQN5GwFKT02fRg5HTUZZHAGLDoidSFNGRk+phJ0UUVkBQ01CREwBhxyaHz1PlZeZA4KdExG4qYOdEs6oqqydsEWphg5/nQ4V1MIDHx0QcsRPxeboA14A
- # [20:52] <cbiesinger> and behave according to what fantasai suggested, I think
- # [21:14] * Joins: Florian (~Florian@public.cloak)
- # [21:17] * Joins: dbaron (~dbaron@public.cloak)
- # [21:23] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [22:08] * Quits: zcorpan (~zcorpan@public.cloak) (Client closed connection)
- # [22:38] * Joins: jdaggett (~jdaggett@public.cloak)
- # [23:16] * Joins: Florian (~Florian@public.cloak)
- # [23:16] * Quits: adenilson (~anonymous@public.cloak) (adenilson)
- # [23:17] * Joins: tantek (~tantek@public.cloak)
- # [23:23] * Quits: Florian (~Florian@public.cloak) (Ping timeout: 180 seconds)
- # [23:34] * Quits: dauwhe (~dauwhe@public.cloak) (Client closed connection)
- # [23:36] * Quits: lajava (~javi@public.cloak) (Ping timeout: 180 seconds)
- # [23:48] * Quits: plh (plehegar@public.cloak) ("Leaving")
- # [23:57] * Quits: darktears (~darktears@public.cloak) ("Linkinus - http://linkinus.com")
- # Session Close: Sat Oct 17 00:00:00 2015
Previous day, Next day
Think these logs are useful? Then please donate to show your gratitude (and keep them up, of course). Thanks! — Krijn