Rust Programming Language Language Design Team, 30 May 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Lang Team Meeting 2019.05.30

Description

Focus of the meeting: triage

A

Go okay! Welcome to this 191 20 1905 30.

A

Name, what's that okay yeah, so we were gonna talk about later today about Josh, can't make it actually in messed up he's calendar. So in fact we're not, but we have lots of other things to talk about what were you going to start with a rep or aligning them yeah.

B

This one Scott just checked their.

C

Box I looks like it should.

D

Be in our CP, no yes, but ours, I think is just slow. But if I say something like piracy, but with you, okay, okay, no! It's done.

D

A

Allow lifetime elision what is happening a.

B

A

B

E

This one again, this.

B

Is such a mess so so self when it contains life times and appears in the the self part of of methods, has different elision rules because it can its anonymously like containing lifetimes, that are part of the self type I.

B

A

Know central, maybe.

B

You can explain the specific thing that's going on in this PR I haven't followed the conversation on this one too enough to understand a pin and mute self specifically is special. My understanding is it's just the only place today where we can have lifetimes in addition to self. Well, it's it.

D

Was special and then I left the Romans, and now it's not special anymore, which means that the title is sort of, so it's the only stable type which you can nest in this way, you can try it box and a sound for a mute so, but you can do it with a with the arbitrary subtypes appreciate.

D

So what I think this PR is doing right now is that it's recursing into alternative parking structures or or other all the places looks like slight expression, if allowed inside his life, and then it's trying to find- and that is trying to find persons that functionally mute X, whereas refers to the cell type, either by writing self or by actually being was outside, like you could rifle instead itself, if you want to, can I just clarify what the behavior is today.

D

That is a really good question like what.

B

Are we the bunker?

B

So there was a there's, a bug before, where a mute self, if it appeared with other lifetimes in the signature, behaved badly with elision, because you you want to have something: that's like an N, mute self method and then it has an N Muse, something you know parameter and then there's an N mu output right. So today on stable that works and that's bad because it's ambiguous which lifetime it's referring to, um and that was done because we skipped the self-type when calculating elisions in some like legacy code that didn't know about arbitrary self-types.

A

So can you make that example of like type it? Are you able to type there's this the example you're talking about right? Not it's probably something else. Right like this as well.

B

So that that X Wilson, this is a particularly an issue specifically with like imple foo.

B

You're doing two space events yeah.

A

That one true Indian, which will never adopt.

A

This one thing.

D

Three diseases, one.

A

Thing I could have changed about rest.

B

B

Yeah this was specific to pin because of the way that it we skipped the self-type. Presumably we make this equivalent to sort of that right.

A

To to this to that yeah, okay, yeah.

B

Right and that's not what we want yeah, it's not sure what should happen. It's really wrong right! That's that's bad, because you would want you would want the changing something from em yourself to pin a new self.

D

B

Initially, a change to undo that entirely and look at the self type. But then that broke other stuff and that's the part where I don't understand. What's what's going on and I think Eddie is Andy Eddie and maybe central or the only ones who do can.

D

We can we not like bound this, so it this happens. We just grow traits like just.

A

A so like, certainly to give ourselves time, we can disable elision right.

D

A

D

Make it because I don't feel comfortable just just accepting this BR without thinking through more okay, so.

B

I mean we certainly can't like just disable elision in arbitrary self-type signatures today, like that's, that's massively stable, creaking rate I'm talking about with when pin appears right, which is that's stable. You can have you can write stable function, I know you're, saying.

A

That it's simply breaking to just disable it with respect to pin, there's enough code correct, okay, right.

B

Like, although all the like futures API stuff is like, do a would be do a unless we like did a bunch of backwards so.

A

There's basically, three scenarios: we could another there if you had more than one reference today like this. It sounds like if I understand correctly, if I had something like, so it would be an error because we're not triggering the like self wins rule we're only triggering the other rule.

A

So really this is dip. So there's there's really. This is the case I'm like basically I'm trying to understand what code would break so this code, but I guess works today as expected. So maybe I'm wrong about this viniq.

B

Are you seen the original issue for where this was brought up? I'm all linkit here, it's it's five to six, seven five I've linked it above I did look at it, but I didn't understand it.

B

So today we skip we, we don't consider alighted lifetimes in the self-type and not in an arbitrary cell type as being like part of the normal allotted lifetimes they're, just like their own independent lifetime, right, okay and and that's buggy, as you can see like from this example, it's just skipping over the cell type.

B

So if you have it in a lighted lifetime that appears in the output, it says: hey, there's no life of corresponding lifetime, an input, and that also means that, like if you have a lifetime, an input and and another argument with a lifetime in the input, it only knows about the second or lifetime argument in the input. So this isn't just a thing for for pinpin just happens to be the only place where anybody actually like really uses this today right. Oh so,.

A

Like you can draw, this is the ready to make an error in an arrow case. Like there's one case of the read the wrong thing: I.

B

Mean it should just do the right, I, don't think we want anything to be an error right, I think we just want it to do the right thing. Maybe we.

E

Want to ignore the pin there, basically, they pretend the pin. Wasn't there because pin doesn't have a lifetime in it.

B

D

I'm not sure I follow that, so so what I, so one thing might do like. If you would do this in a general way, you would be cursed into all type structures from left to right and then just find any any ampersand meat and possibly mutes so yeah.

B

Any syntactically elided lifetime that appears you should yes.

A

It's pretty important that it's syntactic, like that. It's occurring yeah right.

D

B

The right so I think the thing that was going on here right if I'm understanding this the the why the initial fix was broken. It's because self, like it looked at not just syntactically, alighted lifetimes, but any alighted lifetime, which included lifetimes, that appeared in the literal, like self type like when you spells electron little SEL, F and so I think it just needs to be like I think this is just a like.

B

Just a bug quote right like where we need to only care about this, the actual syntactically a lot of life times, rather than the ones that appear underneath the self type or other.

D

B

D

Not clear to me that if you mention and I'm some newts so that that should be considered like if you wrote, I'm son, so I mainly think that we should ever so that we have more.

B

Time to think I think dive into that more actually, because I I would definitely expect that writing in mute and self : and mute capital, SEL F would be equivalent right.

D

So if you, if you are mentioning like is, is the type itself and the trigger here? Is that paper thing or is it? It is I'm son self something special I'm.

B

Not sure I understand what what what's the distinction you're trying make? Are you saying like if I wrote, capital SEL I first? Is writing the name of my type like if you write a few writes.

D

Do do we consider like.

D

Assigned special meaning to understand self, when, when you haven't written any face like you could possibly know so, I.

B

Think those two things should be the same: I think that here, let's let's make this I, think that.

A

That not only should they be the same, but that going back a very long time they were actually sugar. The same thing in the compiler like, in other words from 1.0 or something like that down I, have to go check. What this.

D

Is this is this is probably not doing? This would be a great constraint in anything, I think it's kind of what I'm what I'm getting at.

A

A

D

I think there so they are, they are in.

D

B

Point is that I think that those two things should be different. That the example of last example, I wrote is different because the the foo with the tick underscore and even like, if we can get rid of this and just imagine them it's a lie, dude right. Those should all be that.

A

There is like a tactically.

B

Appearing alighted life time there, which doesn't appear in self, like the words but self does not is include in a lifetime in the way that I'm thinking, okay,.

D

But, but what if we? What if we remove this and write just and then this, then that those two should be the same. But this is not syntactic like how we don't know just from looking at is that this is the subtype. But we know with with no.

B

D

Do know that that's the self type, but not exactly from this. We only know that it. We know I'm.

B

Not sure, what's the what's, the distinction that you're making here of a like was like.

D

How how far away like.

B

The compiler knows the compiler knows here that what the self type is sure but I'm.

D

Guessing that we are doing this for humans, not for the pilot, because I mean we can do but.

E

The human the humans are, depending, on the self part, a thing weird extra case. Do we do we ever state what the actual rule is supposed to be like thoroughly for this somewhere there's I.

D

E

Comment in the PR about I want to know what the new rule is, because it's clearly whatever the PR is doing, is different from what's in the reference right now, right might be fine, but I, don't know what it is and I'm not sure what it's supposed to be, especially with arbitrary cell types.

A

I'm just doing some testing as an aside, our current behavior matches, Taylor's intuition I think it also matches mine I, don't know the answer to whether we have a good I would like an outcome of this discussion to be a good rule. Okay, I'd, like us to write it in the reference. I think I agree with Taylor that it should be syntactic partly for implementation reasons, but also just because it's.

D

That is it's, in fact, if you can say self called an Amazon for.

B

Self : names you can tighten it I, think I mean syntactic in terms of whether or not an alighted lifetime appears in the right.

A

Reference on it, so I'm saying referencing, capital, self and referencing foo are potentially distinct things.

D

Types of a human.

B

I think I I don't care about that like. How are you saying it like today that you just can't do that I think you can? Who are you talking to okay? Well, that seems fine them to me. Then I don't understand the current reference.

D

E

Is about which is, of course, not a 38? Oh, it is about the type and not just attacks.

E

So this is I think the biggest problem I'm having with this whole discussion is I, don't know what the rule we even want to be, and if there was the arbitration, self-types RFC ever talked about what the rule should be for arbitrary sub steps.

B

So I'm I'm, actually confused by this. If the receiver has type and self or in mute self, like I feel like that, this could be far more general like the lifetime.

D

B

In any argument, position that is a syntactically alighted lifetime should should follow that rule. If there is a unique syntactically alive good life time in the input position, then then it should be the life time that gets assigned to lighted output parameter or the receiver.

B

Right, yes, I, not sure I followed that the it has to be for the receiver. No.

B

Well, the if no you know I mean you can have a method that takes self and that, like just self by value and then have another alighted argument and then return like the lighted life lifetime. From that argument, right, like that works today,.

B

Right a little lighted like that. Don't have to appear in insult position.

D

B

D

If you do have a self, then.

B

It doesn't matter where there's yourself or not, it matters whether there's a syntax, a single, unique syntactically, alighted lifetime and input position. Well, that's.

E

One of the rules there's another rule for something. Yes,.

B

E

B

Other rule that you're you're picking out multiple.

E

Lifetimes or maybe, if they're self, this is unclear to me, but if it's unambiguous, which one to use it will use the self one.

A

Yes, for why it's worth the following things: I just want to point out that this this line, in particular at least presently compiles which I found it furious.

B

Because take underscore there isn't necessarily take a so it's not necessarily the self type.

A

But my point is that.

D

What happens if you, if you write seven through self golden person under the take underscore zone, that's a pork feel.

E

Like I think underscore but clean you.

A

Try that my point is more that this, this rule sort of gives a precedence to this lifetime on the ampersand and not this one right it's so there is already something, but you can't just say if there's exactly one region that appeared self-type well, if there is one enlighted yeah, but that's that's pretty inconsistent, because now we also allow, for example, this I, don't particularly like, but you.

E

Can all see use lifetime elision awfully and we.

B

E

B

Yeah we do wait like oh, because there's a there's general there's a line in life time and input position and there's one allotted lifetime and output position. You can write.

C

B

Example should not mile today well today.

A

D

I, don't think it shouldn't. Does it say you assigned to the you ate there today in.

A

C

Know that just I would have expected to be waiting.

A

D

A

Mean you know, I think there should be in it personally. I think this should have been an error and the rule should have been as Taylor is saying and I. Think Central Europe is also saying that you can only match in a lighted lifetime in the output with an allied electrum from the input you can't give a name in one place and not the other. We that is not currently the rule I there. My self types are weird like we could. We've been sort of steering towards that direction and I was even talk of deprecating.

A

This pattern, for example, you know I just learned a thing and I'm terrified, yeah I mean it's got. You know. Another case that you can do is something like this.

A

B

I'm surprised that compiles I would have thought that wouldn't have compiled.

A

It seems like we need someone to write out a rule.

E

E

B

I think I think we should immediately start linting against all instances of that yeah.

E

The a lighting.

B

E

Like I would definitely approve a blinking again so.

A

They're special.

E

Self-Rule tailor I.

D

Really don't want to make any rash decisions I just wanted more time. What week.

B

This this, this Zoo in pole, works today. What does this do? Yeah? It.

E

A

E

Self, that's a special centralized, okay.

B

I'm done this makes.

A

Model and I'm very upset I mean the rule is how I think of the rule is you use empathy if it's an absent self or nipson itself? That's the lifetime. You use. Otherwise, you look for a unique lifetime.

E

A

Matches everything? No, but but not this extended variance like it doesn't apply. We should do with pins. Okay,.

B

I misunderstood this issue and I want nothing to do with this.

B

E

Do we have an RFC that says you can use the special self elision with.

E

Arbitrary self types, or is there only RFC for lifetime militias at the.

D

E

Breath there's.

B

Not even a merged RFC for arbitrary self types at all, there's no there's no RFC for this okay.

A

Somebody needs to volunteer to write this rule out and take charge of this. That's what I said.

D

D

B

Think we should not mirror this like lifetimes in self position thing for, pin mute now that I understand what's going on, because I don't think this should exist in the first place and I think we should start lending against it everywhere.

A

I, don't agree fully with that opinion, but possibly same parts of it like this example, it doesn't bother me this example does, but we there has been a proposal once upon a time very cultivates, legal that takes time.

D

Right looking for some.

A

You know I, don't.

D

A

Who is the person who's going to sort of do the research here? That's what I'm asking you Thanks! That's I'm, not very happy about it. Alright I will do the research. How hard can this be I'll talk to a DBA and look at the code and try to make some kind of yeah.

D

A

D

Before we move on to sterilise on historical stakes, I just want to adventure that I were to construct a order with unsafe facing, so that it is facing. Is that how that works? I just want to mention it to be basic, so.

A

Wait what or you're talking about something it would be different right, yeah.

D

The old sport all right, this.

A

Thing yeah there's some precedent for why unsink she goes unsafe. She goes later.

D

Yeah, it's comes unsafe, unstable, not I'll, say good seems like pretty clear, then yep um so I'm taking this so.

A

What is this stabilize underscore constant in one Greece I.

D

Don't know yeah, we should probably scroll down thanks.

A

For that thought, nothing that's why I was confused. Is that the rest, this one we can defer I think never typed well, can defer that look trying to prioritize here. Let's talk about this one lots of stuff: you go on it central, so.

D

This is the RFC for writing. Cones underscore or type an expression. He comes from. Friday I have written a mostly dumb stimulation reports.

D

But I sort of when I have written most of the report, I sort of last one out I- think there's probably something there which is not on their score and then stuck inside.

D

D

One of the things we already allowance use password, so we could also allow moms from school and stuff inside that would marginally generate humor sunkens in macros. But the drawback of this is that it's not an expression, so you would generate more tokens for macros that need like concert.

D

But mod underscore would keep that in for for driving, macros I think so it's probably ones between which one do you think is better, and then there is so monos what probably doesn't have an engine on organizations which creative thing started. So it's sort of finished after that and whereas cons underscore could be extended to patterns, others then there's questions of what do we do in with those and when they are associated like items and do we want to extend them to hello at any arbitrary refutable patterns and and what about static.

D

So that's something I think.

A

So mod underscore would be some kind of anonymous module, but the idea yeah.

A

Why would we call it model and there's no much smarter anyway, so the well, so that would mean you'd have to it's a bit different than used few eyes underscore I guess like there's, certainly different in purpose, but you would also need to like use super. As I wrote there right.

E

Sorry, if you have mod underscore, is that an a module whose name you can't mention or is it like an automatically used all the insides, or can you only put impulse in it or.

D

Some mother with another name, you confidentially its use.

E

Anything from inside of it.

A

Would be that you can make like, instead of console underscore I, can make like sort of this.

A

E

Sundar score: you have a named Constance ID and unnamed module.

D

You know if you need an expression. Yes,.

D

B

Why would you put a nimble in there because that's so, it's not offer impulse it's for like functions and things that need to type track for the purpose of having a compile time. Assertions on macros.

B

Just that I know about.

D

Well, it's also useful for.

B

Right but that's that's for this- for for those assertions.

D

B

Am fine with mod underscore but I think that we shouldn't block console underscore on this, because I think it's fine.

A

I don't know they both feel like a weird hack to me and.

A

Without like clear benefits but I.

D

Think that constipation isn't happy if you generalize it well, I.

D

Think that would be ice cream and.

E

Maybe in the parts of the Khans fund that returned a tuple or something I could see it yeah.

D

D

E

Could see it, it's not insane.

A

Cuz we just have like I guess, that's the same thing as an anonymous module. It's gonna say just some anonymous. Just clenched there's some text with some stuff. So.

D

I think cons, constant and followed by a block makes sense as as an expression form but I, don't think it makes a lot of sense as like, as a nice yeah. That's fine.

A

I, don't care very much but I feel like it's, not I, guess that you're I think, what's bothering you central is the sort of incompleteness of constant underscore right, the lack of generalization to cons patterns and the fact that static may or may not want.

D

It's not just associated items, it's feels like cattle words, but on the other hand it might be better for, like it works, it works better for the constant sections. So that's maybe like I'm unsure doesn't mean if anyone has a good rationale. I can use. That would be really nice because I reckon so.

B

Central I missed. What exactly are you asking for rush canal for, like.

D

A good reason why underscore constants core is the better choice instead of mod underscore I.

B

Thought we were just talking about how its generalizable to patterns, and then there's also the point about that. We have an RFC for it right.

A

B

It's implemented you.

D

Have implementation takes like a few hours.

E

D

E

Happened since the RFC did something change, you.

B

E

An RFC we had all these discussions in writing new arrives. We.

D

B

I remember seeing things about daughter score before it doesn't. It doesn't feel like a new point to me so.

D

hmm Maybe the do so does this point make sense cause on this floor except accepts an expression on the right-hand side, which means that it's more convenient for macros, then what then not on the school does that make sense.

B

Yeah I mean an alternative, is that you'd have to put a function inside of it, so it's vaguely more convenient, I guess, but doesn't really I, don't know that wouldn't really strongly sway me either way: I guess: okay,.

D

E

You might so maybe not there's no reason to put patterns in module names, because that makes no sense but patterns and konstanz does matter. Yes, there you go.

B

Also have like use name as underscore though or use path is underscore, so there is kind of a weird for a totally different.

A

D

Is okay, I think well, I think I have enough materials, so I will make this and go on rising limb first and you will, from we have before done in a few hours. Alright.

A

Then so, which thing should we discuss next? So there's I'll just run down? We did allow life television, there's some data from NOL, but I think that's not a good choice, because I don't want to tell the others. Well I'll, just my great-great sis we expected probably add.

D

An action I thought an action item to the feeling stuff. Sorry things like when stuff or anything I did that already.

A

Well, I put it under last week's action items, but anyway.

A

So there's stabilized never type again, there's.

A

This type, alias enum variants, discussion.

A

There's so I'm sort of think some of these nominated I'm kind of we could depend a little time discussing these three nominated RFC's and what we plan to do about them. This one I'm going to talk to boats.

D

So I would I would like it if we can discuss like area so I can rise, Newport all right. Let's do that.

D

So do you, people know what this is. I thought.

B

That we said that we were closing this I.

A

B

A which thing we're referring to.

E

The link to it is going to the wrong place by the way.

D

The thing on the screen, perhaps the link- is it's a three yeah the link.

A

Is definitely not the right age for 960? That's fine, I'll fix the loose I, see what.

D

D

Like it, I got there.

A

Okay, so it's this issue where you can access unit, variance to a type alias instead of only through the enum, the actual name so.

D

I mean you can write self variant, which is really nice.

B

Except here in the example, the the type that you're constructing isn't actually at the self type. Oh yeah.

A

This particular example is an error and illustrates one of the interesting side effects of this that you, when you go through a type you are specifying it's quite parameters sometimes and.

D

Yeah, that's what I would expect here. So there was some kind of future compatibility.

A

Thing that came up yes,.

D

So, to summarize, if we have an email with variant and the trait with an Associated life, also named variants, if you then project with so variance, then at the moment it would refer to the Associated type. But if we accept the site there sorry the enum variant types RZ, then we would suddenly had the variant as into the parking space. And what we want petrotech implemented is that's, because environments are inherent, think things. It would resolve with high priorities and beam invariance, and then you would get breakage.

D

So we have a fusion morning that is set to now, so that people don't do this and so that we we can do the do the Union by the types our senior prom yeah. That's.

E

Right, you remind me again: do we support inherent associated types? No.

D

C

But we sure get yes, okay,.

E

But we don't support it yet so that doesn't break this and that'll be a problem when we do support that. Well,.

D

In that case, you can just me name your site, because it's then you can just rename the in our insulin fight because it's parents, so it's like you, have control yeah well,.

E

You'll get a you'll get a duplicate name issue if it happened, because the two things in the same namespace that we wouldn't be able to disambiguate in anyway yeah, but no one wants to do that anyway, so, okay, but for now people can do as trait : : variant to get their associated variant type if they really want it. So we should be fine yeah, and this is we have that ability warning, but it's an allowed breakage, so we're just being nice about it.

A

Right theory did: how long is this work.

D

A resolution I.

A

Did we decide to.

D

E

Think it's there now, so it would be you there for two month, three months by the time we stabilized it's.

D

E

D

Table but it's denying on meatza, so it would.

D

Be a one or sorry, it would be denying for six weeks or it gets well. It keeps being keeps being denied faith mom. So we did it.

A

In January.

D

E

Don't know I feel like no one has an Associated type of the same named as they're eating them very really. It.

A

Doesn't seem there was a certain amount of turn that happened. Okay, yeah.

D

We migrated the other things, as people are just not using those crazy.

D

Okay, then I should write the report or how do we feel about this I.

A

I think this is something people are requested for a long time. It seems like we're making the transition I think you should probably write the report I'm fluent it can.

D

D

Or I can do it.

A

D

There's too much texture, it's really hard silly onions in this document.

D

We should probably separate that especially guys I.

A

D

Like heading through something should probably yeah so never type and well and and the other things. What do we want to talk about and the panic indestructible you know seven ones so I.

E

Don't I don't think panic panic since instructors are bad but they're, not a safety issue. They're like a bug.

E

D

Don't need it to be like e one or anything.

E

D

Yeah Matthew is basically working on things, though it's an implementation issue, I guess all uh nominated and.

E

I, don't think there's any nine questions really right. I, don't.

D

A

Like pretty clear what behavior.

E

Option for Adeline dropship run yeah.

A

I, don't think it's any real.

D

Ambiguity it might be um become in the future, but let's, let's, let's up nothing to Matthews work.

A

It's really another.

D

A

Time for it to run I think so.

D

D

A

I think there's one thing I wanted to discuss, but uh well so there we talked about 2582 last time, but it seemed like we didn't quite get to a plan and I was noting this comment by Ralph. There seem to be some disagreement. I guess abound like can we take some action here and then carve out the rest of the plan ahead of time?

A

How much do we need to like introducing aunt Sandra or something like that and I'm kind of leaning towards that's, probably what we should do, but I I guess I'm, not I'm curious. If people react strongly to that idea, I want to have.

E

Introduced ampersand bra.

A

Yes and trying to sort of committing to that as a syntax, Bend, entre figuring out that will shape the rest of the story around it I. It does feel like it's a potentially large change in some sense for unsafe code authors I mean it's gonna affect a lot of code. I think the rule is probably gonna wind up being. You should just use this like by default. When you run a row pointer, um even though there are only a limited cases set of cases where it actually matters.

E

Yeah I mean it's kind of nice to just get the raw pointer instead of having to then cast it to the rough lender. Any way I.

E

Could make it an arete achill, unsafe ergonomics point it's like if what I wanted was a pointer, an operator that just gives me the pointer is much nicer than the operator. That gives me a thing that I have to cast into a pointer anyway.

D

Right, it seems.

D

I still need to, we check whether we can on ampersand wrong without anything following it.

D

A

Was is that a question of like can as in? Does it affect code and practice or figures? It's clearly. It's.

D

Purely ambiguous in theory, but possibly not in practice, so we could do it. Crazier element see.

D

E

Yeah ampersand raw plus something else or ampersand wrong. Why is theoretically ambiguous but.

D

If you, if you just have an for some cons, then it's absolutely not ambiguous, but we can absolutely do that. The question is: if we remove the cost, can we do that that we have to create a round up to see if it's possible? Oh.

E

E

There's some other word: we could use that we could turn into a real keyword following our normal, like we don't like contextual, key words and expressions things that we talked about around the Edition time right.

D

We do the reveal.

E

Scary, to make.

D

E

Unsafe but that's abuse.

A

E

A

D

We already use Mojang hangings and libraries, no I.

A

Mean there's also yeah okay, so it seems, like we've got some kind of consensus that we should do some sort of syntax it just we're not really happy with the choices at the moment.

D

There is one semantic question which is: if you, if, if we were able to do Amazon draw, should that be for the same mascots or should infer the mutability.

A

I'm not convinced I, I sort of like well I, guess I'm, not sure what I think about that. I feel like I'd, be surprised if he referred to me ability give.

E

You a non of it, could give you a corporate or non-null.

E

Because then it's not meat or cost it's the weird non-null type that it doesn't have two variants just.

A

Get rid of start constant start. You just have star I.

A

Feel like we should do something, but.

E

Nollie is right: yeah.

A

Pretty much no.

E

Is the point order, doesn't have a contour and you know.

A

A

Don't know it seems like we won't resolve it in the next meeting, so I'm going to wait, probably going to try to change the standard, aware cargo, which was next week to be the week after, so that we can do this see parity things I think Josh was wanting to do that more eagerly and then I could try to schedule this pointer, metadata and V table for the week after that, I.

D

Think we need to do something about the infiltrates. Stop because I mean it's is it's important for various things on? Isn't it important for these ways as well I.

A

Agree: I, don't know if we're gonna need a whole meeting for it, but I'll figure that out and and if so I will get it on this counter. Has it that's.

D

A

All right, thanks y'all, you results in things, sort of I- guess not really a lot of time. Talking over luck to me. Oh, it's.

E

Fine, well, we now know that no one knows what lifetime religion is supposed to be that's good. There.

A

You go, we need to speck all right, ciao.