June 8, 2025 • 46 mins
Enjoying the show? Hating the show? Want to let us know either way? Text us!
It's that time of the year again...WWDC!
Last year's conference was dominated by Apple Intelligence...which never really materialized...what's this year's conference going to bring us?
Rob, PJ, and our new co-host, Dave Lewanda (bio episode coming soon) get together to discuss what will be this year's focus.
Will we finally get an answer around Apple Intelligence?
Are there new features and/or hardware around the bend?
Could this "Developer Conference" actually be something that "Developers" want?
All of this and more in the latest episode of Tricky Bits!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
PJ (00:09):
Welcome back to Tricky Bits with Rob and PJ and Dave.
So it's that time of the yearagain, folks, and no, I'm not
talking about the end of theyear holidays.
I'm talking about WW DC, andlet's recap what has happened
over the last two years.
Two years ago, what was the newhotness?
Everyone was talking about theApple Vision Pro.
(00:31):
Last year everyone was talkingabout a different a thing,
artificial intelligence, or asthey like to rebrand it, apple
Intelligence.
And now we're teeing up WW DConce again.
So we have recorded an entireepisode about where Apple
Intelligence is at, and it'snowhere.
(00:52):
What is WW DC going to bring usthis time?
And Dave, this time we're gonnaturn it over to you.
We think you may have researchedthis a little bit more than we
have, so.
are your thoughts?
What does WW DC bring us in2025?
Dave Lewanda (01:11):
Yeah, no pressure for a go with this, but, uh,
yeah, no, I think there's, it'san interesting time, especially
coming on the heels of Google ioand Google really going loud and
proud with their AI stuff.
I think Apple is going to put a,a squelch on it for a year and
really let, let it soak, notdraw any attention to it, and
(01:32):
try and focus attention on otherthings and everyone's gonna be
asking the big question.
Why are you not talking aboutai?
And they're gonna have to dosome jazz hands and, uh, you
know, uh, distraction, you know,you know, close up magic to get
people not to talk about ai.
Um, and obviously it's a biggermiss than something like the air
power pot pad that they tried tolaunch a couple years ago.
(01:53):
Um, even bigger probably thanthe Vision Pro, uh, that didn't
necessarily hit the mark thatpresumably they were looking for
spent all the time spent inpaying really smart people like
Rob to build a piece ofhardware.
Uh, I think this one's gonna bean easier one to bury and then
leak it back out slowly over thenext year and maybe make another
Rob (02:13):
See, I don't agree.
I think they've made the bed andthey have to go lay in it.
And they're still doing thingslike after Google io, there was
reports that Apple may sign up,Google Gemini to be part of the
Apple intelligence backendplatform thing.
And is that gonna be announced?
I don't think they're going toback down on this AI thing.
(02:35):
They sold phones.
We had a podcast on this wholething.
They sold phones based on thisproduct and they prevented that
product from working on olderphones.
And people went out and suckedit up and bought these phones
and it's not here.
And now there's be a new phoneout.
(02:57):
What's the marketing play gonnabe for that?
Dave Lewanda (02:59):
Yeah, no, I think, I mean, everything I've read has
been, you know, they're tryingto again, you know, run some
interference across that, youknow, I wouldn't be surprised
they let you, you know, decidewhether you want to use chat GPT
or Gemini or Claude.
They'll touch button the edges,but apparently the big push is
gonna be unifying the oss, ofiOS 19, it's gonna be iOS 26,
(03:19):
but also MAC OS 26 and
Rob (03:22):
But do we need a conference for that?
That's just a, that's just,let's change it.
It's like that's a typical Applething.
There'll be a day long videopresentation because they've
revolutionized the way thatpeople account for versions.
Even though Samsung's been doingit for years.
it's gonna be groundbreaking andit's gonna be, yeah, we're gonna
(03:44):
name the product after the year.
I think I need a raise for that.
This is the sort of thing youput in like.
Comedy Movers.
Dave Lewanda (03:53):
They're gonna steal the car dealer, you know,
the, the car manufacturer and gowith, you know, next year
apparently it's gonna be 26, noteven 25.
Um, but the other thing isthey're supposed to be unifying,
it's the biggest refresh sinceiOS seven, which is when they
ditched the skew morphicapproach for the flat.
and now everything is going tobe more, the model of Vision os
and all the, you know,semi-transparent translucent
(04:15):
glass.
Trying to bring that look andfeel.
So even to, you know, maybe TVos and watch os get long, do
updates to the appearance.
Rob (04:24):
this sound like a Windows eight to you?
It does.
To me, it sounds like go, oh,everywhere.
This is the new thing andeverything has to have the same
ui, so we're gonna do iteverywhere.
It, it literally sounds likeWindows eight.
I.
Dave Lewanda (04:37):
almost like, you know, they need to run some
interference and buy some timeto get through to 20, 26 or 20,
you know, when piece is a little
Rob (04:45):
Th that is actually a genius plan.
Make something super shitty sopeople get pissed off about it
and they'll forget about Appleintelligence.
If you make something good,they'll be like, oh, it's really
good.
But what about Appleintelligence?
We expect it to be good.
If Apple can misstep and makesomething super shit, it'll buy
that year.
(05:05):
You talked about.
Dave Lewanda (05:06):
up some oxygen, you know, and distract, you
know, run some interference orsome jazz hands, you know, don't
man the man behind the curtainkind of thing.
and you know, it feels, it'sfunny you guys pointed out how,
you know, just like Microsoftmissed mobile, apple missing ai,
this could be like, you know, itwasn't, you know, part of that
was around Windows eight wasaround that same timeframe as
(05:26):
like Windows attempting to stayrelevant as mobile was blowing
up.
That feels like, again, keepingthat.
Story alive or that analogymoving forward?
History repeats itself, Iwouldn't be surprised if they
try and, you know, bury thelead, you know, and keep the AI
stuff tighter to the vest and,you know, by placing it with
something less technically.
(05:47):
Interesting.
Rob (05:47):
That will play out exactly how you said it will.
They just won't say nothing.
They'll just carry on as if wenever mentioned it and kind of
what Windows did or Microsoftdid with the mobile space.
It's like they kind of pro thatfor a bit and then just kind of
went, Hmm, don't talk aboutthat.
I think Apple with ai, if theydon't talk about it, then I will
100%.
(06:09):
Agree with you that that's whatthey're doing.
They, they're just trying to buytime and they don't really know
what they're doing themselves.
And I don't think they do knowwhat they're doing.
I don't think they've ever knownwhat they're doing.
They landed on a cash cow andhaving been in there, I can say
this for a fact, they landed ona cash cow and they have to just
keep milking it.
They haven't had any goodproduct really for years.
Apple Intelligence is not gonnahelp them to make a better
(06:29):
product.
it's always gonna be a featureon a product.
I think Johnny, ive went toOpenAI to make an AI product.
I don't think Apple are capableof doing that.
Dave Lewanda (06:37):
It is probably not.
I mean, Steve Jobs unfortunatedemise at a too young of an age,
you know, whether he would'vebeen able to, you know, pull
another rabbit out of his hat,we'll never know.
this concept, uh, from historyof, you know, the theory of the
retarding lead, you know, thesun never sat on the British
Empire until it did.
It feels like the sun never saton Apple until it does.
Where, you know, are they gonnabe able to, run another playbook
(06:58):
where they can't bring SteveJobs back, though, you know,
where, you know, apple of thelate eighties, to the late, went
through a 10 year fallow periodwhere, you know, Microsoft
propped them up right then.
The next team came back in underSteve's, guidance and, know,
they decided to make the switchfrom Power PC to Intel, and they
had a really nice run fromwhatever it was, two th you
know, os 10 OSX,
Rob (07:21):
been a while.
I mean, it's, it's been 25 yearsthat they've been solid.
I mean, the iPod.
It is really what cemented theirposition.
Yeah.
It revolutionized how wechanged, how we listened to
music and the ads they had withthe white headphones and that
killed off Sony.
Basically.
They were like, you'd com andthen the Windows version of
iTunes was a massive deal, andthen ultimately, obviously it
(07:44):
all got killed by the phone,blah, blah, blah.
We've done over that.
But can they keep doing this isthe big question and I don't
think they can, I think they'reout of ideas of where to put new
products.
They've got one everywhere.
I guess there's the some newthings.
There's the, there's the new carplay, which I think we talked
about before and said it was aalso on that missing list that
(08:07):
we had there.
It seems to exist.
Maybe they'll announce some morethings about that.
My big question is, are theygonna talk about the store and
the new roles from the store?
Obviously things have reappearedFortnight's Bag, biggest app on
the uh, app store right now,
PJ (08:23):
will,
Rob (08:24):
I.
PJ (08:24):
be presented as if it was their decision and it was a win.
Dave Lewanda (08:28):
yeah.
Or, or they just won't even talkabout it.
And, you know, I mean, they'redoing it because they've been
basically, you know, consideredper, you know, committed perjury
and, uh.
We're forced to do it.
Um, I think it's gonna be, it'sinteresting, you know, I think
we've even seen, you know,there's been a transition over
the last few years where insteadof crushing everything out in
(08:48):
the.zero release of the OSS inSeptember, that they've even
already started phasing out.
Like there've been some moresignificant, you know, I
remember it was, uh, groupFaceTime was supposed to, you
know, didn't come out to likewhatever, dot two or there, even
on the dev side, there was somechanges to Swift UI that came
out in the dot four release thatwere pretty substantial that you
would've, you know, midyear ormuch bigger shakeup than that.
(09:09):
So I part of this move to likerebranding the OS by the year
that, know, maybe they evenstart, you know, start using the
month somehow in there as well.
And
Rob (09:19):
Maybe they could put it after a dot.
So they could do 26 dot four forApril.
I am a genius.
PJ (09:28):
You have that guy a re.
Rob (09:31):
Exactly.
Uh, Paul, if you're listening,you can't have that idea.
Dave Lewanda (09:35):
It's been trademarked patented.
Rob (09:38):
Yeah.
Dave Lewanda (09:39):
Uh, tricky bits exclusive.
Um, no, I think, I mean, therethere's been some softening,
there's been
Rob (09:44):
I.
Dave Lewanda (09:44):
I think there's, I mean, who knows what's going on.
I mean, supply chain, you know,the, the whole political
tariffs.
You know, I've heard things thatlike, know, China's blocking the
people that Apple want to takeoutta China to go to India to
make new devices.
So they probably have biggerfish to fry of just continuing
to produce hardware.
I don't know even know how theycould produce new hardware if
they're having trouble buildingexisting, you know, the next gen
(10:06):
of whatever the existing productlines are.
So, that further cements leaninginto the services and the
software products that don'trequire manufacturing or import
export limitations.
There was some mention I readtoday around, uh, you know,
again, more playing on theedges, like the gestures that
you can do with your AirPods,you know, where you can shake it
off, you know, shake your headno to tell it to stop or, there
(10:27):
was talk about camera controlsperhaps.
so I think they've made somelittle inroads again, as you
guys mentioned, where these,it's not trying to be that super
agent or agentic ai that can beyour Jarvis or, uh, you know,
complete control.
But it's these little thingslike I've noticed when I am, you
know, getting in the car, itsuggests, do you want directions
to home?
know, those are the types ofthings, you know, it's
(10:48):
recognizing the pattern that,oh, you know, we go down to see
my in-laws every weekend andthat, you know, it's Sunday
night, you're probably drivinghome from where, you know, this
place 45 minutes away.
So can they continue?
But that doesn't make for a goodstory that they can sell
probably until maybe they canbuild it back up.
And that's why I think, youknow, given this year they'll,
you know, jazz hands anddistract away from it.
(11:08):
And then next year they'll rollout all these little
improvements up into a, a betterstory.
Once they have a better story.
But you can't, you know, whatthey tried last year was lead
with the big news and then fillin the details.
Um, I think now they'll probablytry and get, grab the, you know,
gather the details before theyrelaunch the bigger story.
PJ (11:23):
Let me ask you guys this one, o obviously, you know,
there's a question of what wethink WW CDC is gonna have.
What would you guys want to haveit to have to make it
worthwhile?
Like what would it be like, Hey,I'm really excited if Apple
would solve this particularproblem.
(11:45):
Like, what would be exciting foryou guys?
Dave Lewanda (11:47):
Where WW DC has lost some of its shine for me is
that it's not really as much ofa developer's conference
anymore.
day really feels more like aproduct marketing.
You know, dog and pony show andyeah, they've got the labs or
the, the videos that you canlearn stuff from, but somehow
getting back in touch.
And, you know, I've heard a lotof, you know, even some of the
bigger name developers or otherpeople in the Apple ecosystem, I
(12:07):
mean, John Gruber tore him a newone regarding, you know, they're
not even gonna have a allow anApple person to speak to Gruber
during his, you know, livepodcast that he does at WWC for
the first time in 10 years.
Marco Armand and a couple ofthese other big Apple platform
developers are kinda like donewith Apple.
Xcode has become a bear, youknow, to use,, whether it's, you
(12:28):
know, the macros in Swift, know,having to be compiled and, you
know, running up your buildtimes.
They just released apre-compiled version of the Ma
Swift syntax that is requiredfor macros.
But it, everyone I've read is,it's been tough to integrate
Swift UI previews fails everyother time for me.
And maybe it's part of the, toyour point, Rob, around the
store of like getting back,recognizing the third party
(12:49):
developers that built the Appstore or the App store was built
on the backs of all thosedevelopers and the profits from
the app store.
At least getting back in touchto the developers and really
deciding are they going to beyou know, work with developers
even up to Epic Games
Rob (13:03):
No,
Dave Lewanda (13:04):
going, or is they just, otherwise they might as
well just close the system andmake everything web apps like,
you know, the
Rob (13:09):
I think they'd like that.
It's Apple have never likeddevelopers.
They've always been a thorn inthe side because we've always
said like Apple liked to be thesmartest people in the room.
And when you bring in developersor experts in their field,
they're not the smartest peoplein the room.
This happens a lot in games andgraphics and things like that,
and it's Apple have never givena shit about their developers at
(13:31):
all.
It's like, this is what you'regoing to get and you'll like it.
And if they cared aboutdevelopers, they would've never
have made swift.
Why'd you force on language onit just for you?
It's like, yeah, it's opensource, but no one's really
going to use it.
'cause it has no benefit that ifI'm gonna learn a new language
today, I'm gonna learn rust notswift, but generally I'm just
gonna stick in the c plus plusworld because that's what apps
(13:52):
are tend to be, at leastplatform apps tend to be written
in.
And if you're not gonna doplatform apps, you may as well
do web apps, in which case youhave the whole ecosystem of web
apps for development.
So I don't think Apple Careabout'em from my point of view
in the graphics world, if they'dmake a native version of vol,
it'd be excellent.
make it C.
So we don't have to useobjective C to call any graphics
(14:14):
API.
Which again, if they cared aboutdevelopers, if they cared about
games as they say they do, theyjust bought a tiny game studio,
then you'd have an interfacethat.
They want not telling them whatthey want.
It's like they already haveplatforms they can publish on.
They don't need the Appleplatform.
If they really wanted them,they'd accommodate them rather
than telling them what they'regonna do.
(14:34):
But Apple, I don't think arecapable of accommodating
anybody.
It's, this is we all thesmartest people in the room.
And that will trip'em up in theend.
PJ (14:41):
These are actually really interesting answers.
'cause I think this is a greatdouble down on the DC in ww.
DC stands for Developerconference it's fascinating
because I, I agree with Dave's,notion that, you know, the
keynote has become basically a,a product dog and pony show
Rob (15:02):
but before you go any further, that's true for GDC
too.
These days it seems like alldevelopment conferences have
become whizzbang, flashy thingsand don't, don't have any actual
content.
GDC still has some content, butit's getting down that same
path.
WW DC, the Microsoft one and theGoogle one and the Facebook one,
and they were all just flashymarketing events of, look, we
(15:25):
have developers.
This the days of just a backroom of developers sitting in a
room and hashing it out arepretty much gone.
And that may be because ofYouTube and the way we consume
media today.
It's, and all the informationyou want you can find online.
They don't really need to sityou there.
And secondly, if you go into ameeting with these people, you
(15:45):
can get their info and theydon't want your content in them
directly.
They want you to go throughproper channels, blah, blah,
blah.
There's many reasons why thedeveloper conference has kind of
faded and it's become thiswhizzbang thing, but I think ww
DC is the worst offender.
Maybe Google IO is actuallyworse, but it, it's close.
PJ (16:04):
bad up there.
Yeah.
So, so let me then pivot thequestion.
Do you guys feel, and maybe theanswer's an easy yes.
That these conferences nowoccupy the space as conferences,
really for the shareholders andthe news media and the product
people that want to see the newwhizzbang features, rather than
(16:25):
being about how do we empowerdevelopers to operate better on
our platforms.
Rob (16:31):
Yes, it is exactly that.
I think, and you can see thisbecause it's always from their
point of view, if like Applewill tell you this is how the
new API will be used.
But if you go to like a realdeveloper conference, like I
know the PlayStation's GTC oruh, the internal one for first
party.
Studios, they'll get developersup there saying, this is what we
did.
(16:51):
We told it wouldn't work.
We did it this way.
These are the issues we had.
This is how we worked around it.
The end result is this andthings like, this is how we got
Nan night and visibility buffersand all these new technology.
Was someone just going, what ifI did this?
And it's like, oh crap, it worksreally well.
You don't get that opportunityon Apple platforms.
(17:11):
I will still, I'll go on a stageand say Microsoft's developer
support to this day is athousand times better than
Apple's and'cause they don'treally care about their
developers.
Where Microsoft document all theAPIs, they are very good at like
the MSDN old days and then MSDNonline now it's, it's orders of
magnitude better.
(17:32):
Apple is just literally, they'lldocument the function calls and
it's like, I can read that fromthe damn header file.
Dave Lewanda (17:38):
Yeah, well that's, that's one of the funny things
is some of the documentation,like maybe that would be better
if they just canceled thedeveloper's conference and put
all the money and resources intothe documentation.
'cause often you'll find stuffthat's not in the documentation,
that's in a video, like a ww DCvideo that gets published, you
know, some Edge Corner case forsomething I found, you know, uh,
how you control the center stagecamera on an iPad was not at all
(18:00):
documented in the, traditionaldeveloper documentation, but the
guy who did the presentationmentioned it and, and had it on
one slide in a developer video.
So, yeah, I mean, I don't knowif there's, I.
There's a way, you know, if youwanna have a dog and pony show
for the product team, just do itand call it that launch the iPad
in, June, if you're gonna do thephone and the watch in, uh,
(18:21):
September and the computer, youknow, the Mac in November.
yeah, it really has felt like,it just doesn't, it doesn't line
up with filling the needs of thedevelopers necessarily, but
maybe that's it.
'cause Apple, I mean, that's, toyour point, Rob, I mean,
Microsoft has always been aboutselling their software and
they'd be, would've never madeit anywhere without the third
party developers.
Whereas Apple just wanted tosell you hardware and now it's
hardware and their services.
If, you know, you sign up forApple Fitness and Apple TV and
(18:42):
Apple News and Apple iCloud Plusor whatever and never use a
third party app, they'd be,they'd be fine with it.
Right.
Rob (18:49):
going back to your question pj, what would I like to see?
I'd like to see Apple open uptheir platform of like, these
are all the internal APIs thatwe use.
You can use'em to, that ain'tgonna happen.
And I think if that did happen,you'd see how insecure the Apple
platforms were.
Uh, which is kind of why theykeep doing these wall gardens,
even as the.
And Microsoft does that to someextent, but they are pretty good
(19:10):
at documenting or at leastacknowledging that these are
there.
They say, they don't say youcan't run them, they just say
you probably shouldn't.
and again, good things have comefrom developers doing that of
like, oh, you said don't callit, but we have to,'cause we
have to do A, B, C, whatever ourproduct requirement is.
The requirements of your productis more important than Apple
saying you can or can't do this.
(19:30):
But on.
OSX.
Sometimes you just can't do it.
It's like you absolutely can't,especially if your app's gonna
go in the app store.
'cause then they will vet thatyou're not calling any of these
internal things and you have tonow go via alternate
distribution paths, which forMac you can still do, but iPhone
you cannot do.
So that's what I'd like to say.
Open up and just be more honestas to how these things are
(19:51):
implemented.
And you're not the smallestpeople in the room.
Developers can help you.
They will help you.
They'll tell you better ways ofusing these APIs, better ways of
structuring them.
But you have to be willing tolisten.
And I'd also like to see sideloading on the iPhone.
Get rid of the$99 developer fee,which if you could side load,
you have to get rid of that feebecause why?
(20:13):
Why would anybody pay?
So there's all lots of thingsthat fit together, which just
everything that sucks aboutbeing a developer on the Apple
platform, I think we just talkedabout.
PJ (20:21):
To round back to the same question you, Dave, what are the
things, what are the tech.
Technical things.
I mean, you talked aboutdocumentation, but what are the
technical things you would loveto have coming out of Apple that
they would announce at W-D-W-D-Cthat you'd be like, yes, that is
awesome.
That is worth it.
Dave Lewanda (20:39):
I think just modernizing their APIs., The
mixture of, callbacks and, thereactive approach was hot for a
couple years and then, you know,now you know, async weight, you
know, mean swift.
I liked Swift.
I got into Swift.
I, I thought I had a lot ofSyntactical cleanup, you know,
more concise, you know, andeasier to read, easier to teach.
And I think that's kinda wherethey were going.
But it up through Swift five,like Swift 5.9.
(21:02):
It of a couple years ago, andthen they started to move
towards Swift six, which is thebig move to the concurrency
stuff.
It all started to fall apart andit's become a, a mess.
And then the fact that like alot of the Apple APIs,
especially the Apple frameworks,are just a mix.
They, they haven't bothered, youknow, the, uh, keeping them at
all in sync.
You know, each one is slightlydifferent, different generations
of callbacks or key valueobservation or having a
(21:24):
publisher interface and nothaving any level of consistency
there.
So even if you wanna structureyour app, you end up having to
wrap it, your, each one, eachone of their SDKs, whether it's
core Bluetooth, whether it's AVfoundation in a way that works,
to build your own.
Consistency on top of that,which makes, you know, the
mobile app of any, any level ofcomplexity that much more
complicated.
(21:45):
'cause all those things are sortof, can move underneath the hood
for you.
As we, as I dealt with in, youknow, jobs.
There's gotta be an answersomewhere here around, in the
blog post I wrote up aroundhaving a single platform for
these mobile devices, you know,I don't think it's React native.
I, I'm not sure it's flutter.
There's been a big push for,there's a Swift on Android group
and maybe they'll talk if theytalk about that at wwc.
(22:05):
Um, apparently they just did abig.
post, I haven't had a chance toread it yet on how they use
Swift to rewrite the swift, onthe server side for the
monitoring for the ApplePasswords app.
the Mac, you know, the AppleiCloud passwords app that they
pulled out of the OS and made astandalone app, probably because
of competitive, concerns, withthird party vendors like One
Password or LastPass., So Ithink there's opportunities
(22:27):
when, when Swift is used, well,it feels like it can be useful
and, you know, can be veryexpressive.
I've, I've enjoyed that, but italso can get very messy and,
know, just be a, a dog on a lotof these things because
especially you know, theconcurrency problem, they try to
solve a lot of the reallychallenging stuff, but messed up
the easy part.
They pushed everybody into thedeep end of the pool, but before
(22:48):
everybody was ready to, and nowthey're that, and Swift 6.1, 6.2
is tried, but I haven't, and Ihaven't been as close to the
code in the last few monthsmyself, but everything I've been
following online suggests thatit's, you know, they've, it gets
messier before it's gonna get
Rob (23:01):
Isn't that a symptom of being the smartest person in the
room?
Dave Lewanda (23:05):
Well part of it, the, the guy Chris, uh, Latner,
I think was his name, I'm mightbe mispronouncing that, who, was
the, one of the originalfounders of Swift Left Apple.
And I think that's sort of when,you know, he's the one who put
together the originalconcurrency pitch.
similar now, to, Steve Dobbs,when who take over for the
vision lose sight of the, thereasons and the, the motivation,
you know, they're trying andthey're probably not, you know,
(23:26):
I don't, again, don't wanna bessmart to anybody, a lot of smart
people who are doing it, but,you know, maybe they're
overthinking it.
I don't, I don't know.
It's a, it's clearly a hardproblem to solve,
Rob (23:34):
It is, but they never do it in like a layered approach.
Like you could just be like, ifyou took the C approach, even
let's just take threads in theoperating system as we know'em
today, they haven't changed in50 years what they are.
That's a thread.
If you want to do asynchronousat some point you create one of
these and, and then on top ofthat, you build various.
(23:54):
Asynchronous models and whohandles creating and destroying
the thread, who handlessynchronizing them, who handles,
when they run and when theydon't run.
Things like that, that can allbe in the higher level
frameworks.
And even Swift, it's in its highlevelness, it's still doing this
at the operating system level.
So I think the whole vision oflike, any of these layers you
(24:15):
can interject at based on whatyou want for your app.
Apple never do that.
I, I'd love to see them becomemore of a, a straight up.
Like this is the platform, theseare the levels and layers that
we have.
And comes back to what I saidabout Volcan Avenue C interface.
it'll never happen, but it'd benice to do that.
And I still think a lot of theold fashioned platforms like WIN
(24:36):
32 or X 11 or things like thatare still have that thing.
You can use these big fancyframeworks on top.
But generally it's built on thisand it's that this bit that's
missing on the Apple platforms.
And I've always thought thathaving to throw in a new
language and a new model and anew concurrency model and all
that on top of everything elseis just, I think it's hard.
(24:57):
I think Apple's a very hardplatform to develop on.
Xcode you mentioned is horrible.
It can't even give me errors ina file in order it has to order
them.
For me, it's like, why can't youcompile like every fucking other
platform compiles of justcompile a file, tell me what's
wrong with it.
Dave Lewanda (25:12):
best one.
If, and not that you needed anyre convincing that swifty y is
bad, but right now it'll justsay it can't compile in a
reasonable amount of time andyou end up having to do a binary
search by like commenting outthe whole block and then
uncommon half of it, see if itcompiles, and if it tells you
what the real error is.
Nope.
Comment that back out uncommon,like that feels like a just
complete miss.
As far as the tooling on thetooling side,
Rob (25:32):
Yeah.
And annoying things like thatwindow with the errors in, it'll
scroll up.
there's so many inated areaslike show if you know this one
caused it, pull it at the bottomso I, or put it wherever the
focus of the window's gonna be.
it's like, don't make me scrollall the way back up and do it,
at least to the CI mean, ithappens to see too, but at least
it's like file by file.
Like, I know these are all goodand this one isn't.
And it's just ridiculous thatthey do things so different to
(25:54):
everybody else where there's noneed to
Dave Lewanda (25:56):
I mean, I think going back to your question,
like what would I want if, if Icould have one wish, you know,
the genie came outta the bottleand I had one wish for the Apple
Developer Conference.
Fix X code, just make X code
Rob (26:08):
make Xcode Visual Studio.
'cause that's the, that's theone thing that Microsoft's
always made.
Well, I was, I've never been ahuge, huge fan of it.
Mostly because it has a terribledebugger, but compared to the
debugger in X code.
Visual studio might as well besome like kale embedded
debugger.
It's, it's great.
Dave Lewanda (26:27):
I'll, tell you still my favorite ID of all
time.
Uh, and even with all the workthat's gone into like the AI
enabled stuff not that I'vespent a ton of time with, that
is still net beams that sun putout for Java swing development
in 2006.
It was amazing.
It would, you know, couldrefactor code every, every which
way you could almost to thepoint, like similar to what you
(26:48):
see some of the LLM stuff doingnow.
It's like, I need a function.
It would fill in, it wouldautomatically fill in blocks
and, you know, help you, youknow, complete the code in a way
and point out where the bugs,the errors were.
I mean, you'd never have acompiler.
It almost had, it felt like aspell checker, like as you were
typing it would, you know, getyour syntax and tell you the
wrong thing.
And
Rob (27:05):
See, for me it's not about the idea, it's about the
debugger.
It's like if you went in lowlevel code or very technical
code, I just wanna step throughthe function and see what it's
doing.
And I think the debugger is thekey.
And for me, SN Debugger for thePlayStation was brilliant.
And then Sony killed that'causethey integrated everything into
Visual Studio and it's gettinggood again now.
(27:27):
But it's took them well 15 yearssince the 20 years since they
killed the SND bugger.
Uh, I think the KLD bugger, thearmed debugger is really good.
Sega Ozone Debugger also anembedded debugger.
Really good.
Uh, mostly'cause a lot of theseare scriptable debuggers.
So you can write scripts to dothings that don't have function
(27:49):
native functionality within thedebugger.
Technically excode can do it'cause you can do python in
A-L-L-D-B, but uh, it's soobfuscated that it's an exercise
in frustration just to get toit.
Excos always been the worst ofall the ideas for sure.
but it's interesting'cause allthe things we're talking about
should be things that are at adeveloper conference.
(28:09):
Enough.
Nothing that we just talkedabout for the last hour will
even be relevant at ww DC'causeof the, the whizzbang and the
flashes that they need to putout there, as PJ said to, to
keep the shareholders happy morethan anything.
And we are doing something, weare making progress.
Dave Lewanda (28:24):
not the the old crusty guys like us, they're
trying to get, you know, newpeople, you know, a more diverse
developer crew that a lot of thestuff is, you know, geared
towards the student developer orthe, you know, making it as easy
as possible, lowering the, theon-ramp.
A lot of, you know, the swiftUey stuff was clearly to move in
that
Rob (28:41):
Well, if you go in that direction, do what you said and
do a web app and get it for allplatforms at the same time.
Dave Lewanda (28:46):
Again, they're trying to lower the barrier to
entry to get people, you know,so that, you know, those kids in
20 years are doing a podcastabout how bad Apple is in 2045.
But, who knows what they'rereally thinking?
It's clearly a hard problem.
And, you know, again, I'm nothere to, you know, drag him on
anything, but he certainly coulddo better.
It feels like, it doesn't seemlike it'd be that hard to do
better.
Rob (29:03):
Oh, I'll drag over the cold.
not a problem.
Doing that, having worked there,I.
Internally, it's as messy as youthink it is of like I can tell
you exactly how those frameworkscome become to be so disjoint
and all of that.
And they do have an APIstandards committee inside.
So you have every API you haveto submit to it.
And they do supposedly to dolike style of API auditing and
(29:27):
they also do security auditingand things like this function
could be a problem, but theydon't, they're not by good
because loads of crap gets pastthem.
They have standards like inteshave to be intes.
Like if it's, even if it's anunsigned value, it still has to
be an integer because in 64 B,you're not gonna have a number
that's bigger than you can storein 64 B, but, but things like
that, it's like you're being thesmartest person in the room when
(29:49):
you enforce things like that.
And I've said this exactly asI'm gonna say it again on this
podcast before is.
Unsigned INT is a safer datatype because if you pass in a
negative number, it shows up asa massive positive and the check
you have to say, if X is lessgreater than 10, we'll catch all
negative numbers and only zeroto 10 will get past it.
(30:12):
If you are suddenly forced tohave that value being integer,
because some API standards boardsaid, so you now have to
remember to go back andexplicitly check for all the
negative values and be like, ifX is less than zero, then it's
also an error.
And lots of times and lots ofbuffer overruns and lots of
security problems have comeabout because you can index
(30:33):
negative in this functionbecause no one checked the
negative value.
It's just things like that of,and even in the on go into a
whole different subject on theVision Pro, there was swift
inside system code and there wasa, a system wide, architecture
decision.
Now you can't allocate memory.
Which basically means systemcode has to be C, which is what
(30:53):
it should be, not even c plusplus in the grand scheme of
things, it should be C, the OSis straight C.
The system code should bestraight C as well because you
can control when it allocatesmemory, objective C and Swift
will even objective C will dothis and Swift does it more, but
objective C does it.
When it reallocate its dispatchtables'cause it's like, oh I
need to reallocate the cash bangmemory allocation.
(31:15):
we didn't prevent memoryallocation by saying you can't
call Malo.
We blocked the call.
So to make it even worse is Malowould sometimes work'cause you
go, I need a K and it's got a Kin system in a user space, so
well give it to you and it's allgood.
But then every now and thenMalik's like, well I need
another page from the operatingsystem.
So it'll go and hit the SIScall, then it will crash and it
(31:35):
crashes miles and miles awayfrom the thing that actually
caused the initial problem.
And that's a good thing reallyyou do want, thats call to be
blocked.
But if you're gonna do that,you've gotta structure your code
above it in a way that'sexpecting, thats call to be
blocked and know what things aredoing.
People would just show up and,oh, I'll just write this in
(31:55):
objective C And I've said thisbefore too, we, that API that
would pass a dictionary inbecause people couldn't agree on
what the fucking parameters forthe function code should be.
So we'll pass a dictionary in'cause then you can just query
it and see if your parameter'sthere.
And if it is, you can do this.
And if it isn't, you can read adifferent parameter and all that
of like, how bad do you wantthis to be?
(32:16):
And it's system level code thatjust gets a billion times worse
when you get up to the highlevel code.
They do not have the rightpeople in the right places is
really what it comes down to.
And we've had a podcast on thistoo.
It's.
People get hired and they getmoved around.
And even on my team, there was aguy who was brilliant at math
(32:38):
and all this and he was doingembedded code and I was doing I
level code and we could have alljust took his position to the
left and would've all been in amuch better position to, to do
things.
But not because the head countand how the teams work and all
of this and who needs good staffand who doesn't it, it was just
a typical middle managercorporate mess.
PJ (32:57):
Welcome to Big Tech Folks.
Rob (32:59):
It's a mess, but they, they have this API board to try and
fix it, and they enforce thingslike ins have to be ins and not
uin, and it's like you're notactually fixing anything.
You really have to go back towhat we said about GDC and open
up the layers, clean it all upfrom the ground up, and not
build all this on a basic houseof cards.
(33:20):
It needs to be solid foundation.
Then you can go to the highlevel languages on top of that,
but the foundation has to besolid.
But most people have thatlayering system Windows as if
Linux, as if Apple have it atthe OS level, like Kernel is
clean and fairly good.
Above that, it's fucking freefor all.
Dave Lewanda (33:40):
I wonder if there's a space for, you know,
if that's where AI could beuseful is, you know, turning
what people are asking for intowhat they, the way they should
do it and you know.
Things, from running amuck it'smore regimented and it, you
know, could learn over timemore.
So, you know, like self-driving.
Could you have self programmingwhere you say, I want to get
from, you know, instead of Iwanna go from point A to point
B, I wanna get this piece offunctionality and then have the
(34:03):
AI really enforce more of thestricter rules, more rigidly
than, you know, humans are.
'cause humans will always fightover, you know, the bike
shedding of, you know, un versusin or four, you know, four
spaces versus a tab versus twospaces
Rob (34:18):
Two spaces is the answer.
There's no argument.
It's two spaces.
PJ (34:24):
All right.
That'll be a topic for anotherpodcast.
So for ww DC we would love it tobe about the DC won't be.
It sounds like DC will not standfor de developer conference.
It'll stand for damage controlfor, for AI at, at best, and
maybe a few passing featureshere or there.
(34:45):
Gentlemen, Are there any majorfeatures we expect to see out of
this next conference, even froma product shareholder marketing
standpoint, or is it gonna be ofsmoke and mirrors around minor
stuff like we've seen in thepast?
Rob (35:03):
I don't know.
I, uh, I'd like to see some newhardware, but I don't think even
the rumor mill doesn't reallyhave much new hardware.
A lot of it's been announcedjust along the way, various
releases here and there.
Maybe they're stepping away fromthe, the all in one hardware
announcement.
Maybe there's a new Mac studio,an M four based one, maybe the M
four ultra, whatever the JUULone is.
(35:25):
I think that's the ultra, maybe,I don't know.
But I think if there's anyhardware, it'll be fairly small
things for a refresh.
Dave Lewanda (35:33):
M five chip, you know, because the M four has
been out for, you know,announced that the M five is
coming later this year.
AirPod Pros, maybe I could see,you know, the, now that the
AirPod pros are like, you know,medical equipment and that
they're hearing aidreplacements, they haven't
really touched that.
I would love to see the home podwith a screen, but, you know,
that, or even just better Homepod support, that that's been an
(35:55):
area that they've let languish,you know, I was hoping it'd be
kind of like the Apple tv, youknow, the original Poppy project
as Jobs called it.
And then they turned it intosomething a bit more real over
time that like the home podfeels like it's ready for,
coming to the table as a real,as a real player, again, putting
a screen on.
I, I have a perfect use case inmy fa in my, you know, kitchen.
You know, if I had a screen thatI could put like a family
(36:16):
calendar on you, talk to the,you know, with, especially with
Google stepping away from theirhardware.
More of the home kit space.
Some more support for just smarthome.
But I dunno, that feels likeit's fizzled
Rob (36:25):
I think.
I think that's the solarhardware you're gonna see.
You're gonna see the smallerthings.
There's no new phone.
There's no new I.
There might be a new iPad.
I don't think so.
Maybe a new web of one of them.
I think they've
Dave Lewanda (36:35):
I think, you know,
Rob (36:36):
done them all though.
Dave Lewanda (36:36):
studio display, maybe the next generation studio
Rob (36:39):
A big iMac.
I would like if they'd make a 30inch iMac, I would buy one.
Dave Lewanda (36:44):
that all that's gonna be limited by, you know,
with the political and tariffsand import export challenges.
I, I think maybe that's wherehardware just is DDOA this year.
Rob (36:53):
Maybe.
PJ (36:53):
make a prediction and I, I suspect this is something they
will use the, tariffs as a coverstory for, I think they will
deprecate the Mac Pro.
I think it has existed basicallyfar beyond its useful lifetime
they can use the cover story ofthe political stuff as
effectively a way to get rid ofit.
Rob (37:17):
I think we had a whole podcast on this exact thing as
to hard take on the Mac Pro andI would, I think they announced
they, we said this before I.
They released it because theysaid they would, it's an
abomination of what theyactually released.
You can't put new memory in it.
You, you can put PCI cards init, but you can't put a GPU in
it, blah, blah, blah, blah,blah.
(37:38):
And I could see them just beinglike, yeah, it's, they did it.
They said they do it and that'swhy they did it.
And now it's time to let itmaybe die quietly.
PJ (37:47):
Well,
Rob (37:47):
And
PJ (37:48):
can use, they can use all the political
Rob (37:51):
I think so yeah, it's time to let it die.
It's like, let it die quietly.
And it's.
PJ (37:55):
a new Max studio like you guys are saying at the same
time.
To say this is occupying thishigh-end space now.
Dave Lewanda (38:02):
Yeah, only'cause I've heard, you know, with what
and seen the stuff like maybemeta's done with glasses.
Could they have another visionPro or vision not pro their
sleeve somewhere?
Rob (38:15):
They would, they'd be more leaky things about it if it
happens.
So they were looking at thatsort of thing when I was there
and there is a second visionprobe being made, blah, blah,
blah.
but I don't think we're in amarket space that wants that
product.
and I think meta may be gettingthis more right than Apple is
(38:36):
when you have a new interfaceand you alerted this earlier
with the keyboard mouse going tothe touchscreen and go into the
ai uh, products.
If you have glasses that are onyour head and they're gonna be
there for a long time, you'vegot to ease people into it.
You can't just be like, oh look,here's a full experience.
And I think if we had glassstyle, like meta style glasses
(39:00):
today that were full ARexperiences, people wouldn't use
it.
But.
If you take a small step and be,okay, I'll just put your
notifications in your face soyou can look at them while
you're driving.
I'll pull a camera in it so youcan take, video.
Basically things that meta did,that's the stepping stone that
people need to get used tohaving this thing on their head
(39:23):
all day.
You can't put this thing on yourhead and have this intense
experience and expect to wearthese for 12 hours.
Dave Lewanda (39:29):
That's where I, think maybe again, the AirPods
have a space there to be thatstart being that bridge.
You could, you know, put alittle camera in the, in the
stem somewhere, or eveninfrared, you know, some level
of sensor, you know, if they'redoing, getting you to do
motions, like shake your headno.
Or, you know, those type ofthings.
You know, more, more smarts inthe
Rob (39:46):
yep.
Yeah.
Dave Lewanda (39:46):
again, way not to cover your face, you know, I
mean, a number of folks havepointed out, you know, just
having something on your face asa detractor, know, losing your
peripheral vision and or, youknow, your ability to track
mates a, uh, thing that, youknow, just human evolution has
shied away from.
So, you know, can you do it allfrom your ears?
And maybe it's, you know, biggerear pods or bigger, you know, a
(40:07):
headband that doesn't go overyour face is there, you know,
other places to put hardware,whether it's tying to your phone
in your pocket to your watch onyour wrist, to your, I don't
know what other, where else youcould stick, compute or, you
know, bandwidth that would work.
But, Maybe just not gettingaway, getting away from the
face.
And that's where I think maybethe AirPods have some space to
grow and, and even especiallywith them till off the cans.
(40:30):
'cause they realize the cansaren't a solution, but more
focusing more on the, theearbuds.
Rob (40:36):
I think we missed something else too.
There's some game stuff thatthey're gonna announce.
They have that new games Xboxstyle game center that they're
gonna announce.
Dave Lewanda (40:45):
app,
Rob (40:45):
just replacement for Apple Arcade.
I think it's just, I think abetter way to find them, but I
think it's still, they're notgonna take games seriously.
Dave Lewanda (40:53):
you know, they pull it out of the app store and
make it into a, like, you know,apple News for
Rob (40:58):
Yeah.
Dave Lewanda (40:58):
and Apple Sports.
Rob (40:59):
it'll be.
Yeah.
And I still think, and I'vealways said this, they're
missing a huge opportunitybecause the Apple TV is quite
powerful.
It could do some pretty decentgames if they opened it up
console style to actual consoledevelopers.
Dave Lewanda (41:15):
and they wanted to, with that original remote,
you know, that had the motion,you know, you used to have like
a Wii style remote in that, theone that had, you know, just the
flat
Rob (41:23):
Yeah.
I mean, it'll connect to aPlayStation controller too, of
like, if, I dunno what that saysit though, is that Apple saying,
we can't make a bettercontroller, so we'll just use
yours?
Or is that Apple saying, wedon't give a shit, we're not
gonna put effort into it.
We'll, uh, we'll just use yours.
PJ (41:40):
My suspicion basically is that they won't get into the
game space because of the, thedomino effect, Rob, that you're,
you're kind of alluding to isthat really do it right, you
would need to put Vulcan assomething native on all of these
to do that.
You then say, okay, well thenhow do I open up the OSS in such
(42:04):
a way that everyone can seeeverything?
In which case I think you hitthe security issues.
So I, I suspect that there'sthis cascade of stuff where it's
like, if you really want to dogames right on the Apple tv, on
iOS, on the Mac.
You end up actually opening upthis gigantic can of worms that
Apple simply doesn't
Rob (42:25):
And it's an ecosystem that already exists here.
It's if you use, if you're gonnamake AAA game, you're gonna use
Ws and you're gonna let it dothe mixing, and you're gonna,
you need basically access tocontrols, video and audio.
Network sockets.
And outside of that, everythingelse could be layers.
But now these layers thatalready exist don't fit this new
(42:47):
framework.
So we're saying, I think it maybe part of what you're saying,
Dave Lewanda (42:50):
if not anything, it feels like they tried, they
did that when they did that, youknow, scary event.
It was Halloween a couple yearsago, and it was at like 8:00 PM
Eastern, so it could be on andduring the daytime in Tokyo.
And they had that famousJapanese game designer
Rob (43:03):
they always have a game.
Every single time they release anew GPU, they always have AAA
game, but they've paid them toport that and they jump through
the hoops of calling theobjectives and all that.
So I'm not saying you can't doit, I'm just saying developers
won't do it because it's the oddball.
Dave Lewanda (43:20):
Right, because well, yeah, so I mean, it feels
like if apple apple's dippedtheir toes in all those times
and never done it, they're,they're never going to tip pj.
To your point, they're nevergoing to do it.
PJ (43:29):
I, I just think it takes way more than they're willing to
commit.
you have to start from theperspective that I care about my
developers, in which case youhave to ask your developers,
what do you need?
Rob (43:42):
Yeah, they're capable of doing that.
PJ (43:45):
that step, then again, you can pay people to jump through
your hoops perfectly willing todo that, but you're not gonna
get the, the best, uh, the bestdevelopment experience, the
Rob (43:59):
Games is the extreme example of you can't do it
yourself I mean, so is kind ofgetting there by buying all the
studios, but the fact they ownthem doesn't mean that they have
any say over them.
think PlayStation by itself isuseless.
And we've talked about this likePlayStation will win every
single console war because ofits exclusive content, which
(44:20):
come from developers.
Like, so don't make thatthemselves.
Like I says, they are graduallybuying up all the developers, so
technically do own them, but,uh, it's still independent
developers, independent mindsetof developers at least that
makes these really awesomegames.
And Dave said it earlier ofit's, these are the frameworks
(44:40):
you get and do use it.
And if, if they can give yousome services and give you
something that you'll use as is,then it's, it's what they'll do.
And again, it's the smartestperson in the room syndrome.
It's, it doesn't work for gamedevelopers and there's never
been a platform more extremethan PlayStation for, without
the developers, this platformwon't even exist.
(45:04):
Unless you're going to, I thinkthere's also a gray area, like
what is, we talked about thistoo.
What is the future of consoles?
Maybe it is Apple tv.
Maybe it is like get rid of thebig consoles and just have
something that could stream fromPlayStation Cloud or Xbox Live,
or whatever it is.
So maybe it's not worth goingdown the path, but they already
(45:24):
have the hardware, so it's notlike they're making new hardware
and having to spend a billiondollars promoting it.
They could literally just changea few things and they'd
immediately be a, a verycompetitive gaming platform.
PJ (45:41):
Well, we've got some fun predictions for what Monday may
bring.
I don't think any of us areespecially optimistic at this
point in time, but what we'll dois next episode, we'll do some
grading and some commentaryafter the fact to see exactly
what we got out of this WWDC.
So stay tuned, folks.
(46:02):
This story is not over yet.
Welcome back to Tricky Bits with Rob and PJ and Dave.
So it's that time of the yearagain, folks, and no, I'm not
talking about the end of theyear holidays.
I'm talking about WW DC, andlet's recap what has happened
over the last two years.
Two years ago, what was the newhotness?
Everyone was talking about theApple Vision Pro.
(00:31):
Last year everyone was talkingabout a different a thing,
artificial intelligence, or asthey like to rebrand it, apple
Intelligence.
And now we're teeing up WW DConce again.
So we have recorded an entireepisode about where Apple
Intelligence is at, and it'snowhere.
(00:52):
What is WW DC going to bring usthis time?
And Dave, this time we're gonnaturn it over to you.
We think you may have researchedthis a little bit more than we
have, so.
are your thoughts?
What does WW DC bring us in2025?
Dave Lewanda (01:11):
Yeah, no pressure for a go with this, but, uh,
yeah, no, I think there's, it'san interesting time, especially
coming on the heels of Google ioand Google really going loud and
proud with their AI stuff.
I think Apple is going to put a,a squelch on it for a year and
really let, let it soak, notdraw any attention to it, and
(01:32):
try and focus attention on otherthings and everyone's gonna be
asking the big question.
Why are you not talking aboutai?
And they're gonna have to dosome jazz hands and, uh, you
know, uh, distraction, you know,you know, close up magic to get
people not to talk about ai.
Um, and obviously it's a biggermiss than something like the air
power pot pad that they tried tolaunch a couple years ago.
(01:53):
Um, even bigger probably thanthe Vision Pro, uh, that didn't
necessarily hit the mark thatpresumably they were looking for
spent all the time spent inpaying really smart people like
Rob to build a piece ofhardware.
Uh, I think this one's gonna bean easier one to bury and then
leak it back out slowly over thenext year and maybe make another
Rob (02:13):
See, I don't agree.
I think they've made the bed andthey have to go lay in it.
And they're still doing thingslike after Google io, there was
reports that Apple may sign up,Google Gemini to be part of the
Apple intelligence backendplatform thing.
And is that gonna be announced?
I don't think they're going toback down on this AI thing.
(02:35):
They sold phones.
We had a podcast on this wholething.
They sold phones based on thisproduct and they prevented that
product from working on olderphones.
And people went out and suckedit up and bought these phones
and it's not here.
And now there's be a new phoneout.
(02:57):
What's the marketing play gonnabe for that?
Dave Lewanda (02:59):
Yeah, no, I think, I mean, everything I've read has
been, you know, they're tryingto again, you know, run some
interference across that, youknow, I wouldn't be surprised
they let you, you know, decidewhether you want to use chat GPT
or Gemini or Claude.
They'll touch button the edges,but apparently the big push is
gonna be unifying the oss, ofiOS 19, it's gonna be iOS 26,
(03:19):
but also MAC OS 26 and
Rob (03:22):
But do we need a conference for that?
That's just a, that's just,let's change it.
It's like that's a typical Applething.
There'll be a day long videopresentation because they've
revolutionized the way thatpeople account for versions.
Even though Samsung's been doingit for years.
it's gonna be groundbreaking andit's gonna be, yeah, we're gonna
(03:44):
name the product after the year.
I think I need a raise for that.
This is the sort of thing youput in like.
Comedy Movers.
Dave Lewanda (03:53):
They're gonna steal the car dealer, you know,
the, the car manufacturer and gowith, you know, next year
apparently it's gonna be 26, noteven 25.
Um, but the other thing isthey're supposed to be unifying,
it's the biggest refresh sinceiOS seven, which is when they
ditched the skew morphicapproach for the flat.
and now everything is going tobe more, the model of Vision os
and all the, you know,semi-transparent translucent
(04:15):
glass.
Trying to bring that look andfeel.
So even to, you know, maybe TVos and watch os get long, do
updates to the appearance.
Rob (04:24):
this sound like a Windows eight to you?
It does.
To me, it sounds like go, oh,everywhere.
This is the new thing andeverything has to have the same
ui, so we're gonna do iteverywhere.
It, it literally sounds likeWindows eight.
I.
Dave Lewanda (04:37):
almost like, you know, they need to run some
interference and buy some timeto get through to 20, 26 or 20,
you know, when piece is a little
Rob (04:45):
Th that is actually a genius plan.
Make something super shitty sopeople get pissed off about it
and they'll forget about Appleintelligence.
If you make something good,they'll be like, oh, it's really
good.
But what about Appleintelligence?
We expect it to be good.
If Apple can misstep and makesomething super shit, it'll buy
that year.
(05:05):
You talked about.
Dave Lewanda (05:06):
up some oxygen, you know, and distract, you
know, run some interference orsome jazz hands, you know, don't
man the man behind the curtainkind of thing.
and you know, it feels, it'sfunny you guys pointed out how,
you know, just like Microsoftmissed mobile, apple missing ai,
this could be like, you know, itwasn't, you know, part of that
was around Windows eight wasaround that same timeframe as
(05:26):
like Windows attempting to stayrelevant as mobile was blowing
up.
That feels like, again, keepingthat.
Story alive or that analogymoving forward?
History repeats itself, Iwouldn't be surprised if they
try and, you know, bury thelead, you know, and keep the AI
stuff tighter to the vest and,you know, by placing it with
something less technically.
(05:47):
Interesting.
Rob (05:47):
That will play out exactly how you said it will.
They just won't say nothing.
They'll just carry on as if wenever mentioned it and kind of
what Windows did or Microsoftdid with the mobile space.
It's like they kind of pro thatfor a bit and then just kind of
went, Hmm, don't talk aboutthat.
I think Apple with ai, if theydon't talk about it, then I will
100%.
(06:09):
Agree with you that that's whatthey're doing.
They, they're just trying to buytime and they don't really know
what they're doing themselves.
And I don't think they do knowwhat they're doing.
I don't think they've ever knownwhat they're doing.
They landed on a cash cow andhaving been in there, I can say
this for a fact, they landed ona cash cow and they have to just
keep milking it.
They haven't had any goodproduct really for years.
Apple Intelligence is not gonnahelp them to make a better
(06:29):
product.
it's always gonna be a featureon a product.
I think Johnny, ive went toOpenAI to make an AI product.
I don't think Apple are capableof doing that.
Dave Lewanda (06:37):
It is probably not.
I mean, Steve Jobs unfortunatedemise at a too young of an age,
you know, whether he would'vebeen able to, you know, pull
another rabbit out of his hat,we'll never know.
this concept, uh, from historyof, you know, the theory of the
retarding lead, you know, thesun never sat on the British
Empire until it did.
It feels like the sun never saton Apple until it does.
Where, you know, are they gonnabe able to, run another playbook
(06:58):
where they can't bring SteveJobs back, though, you know,
where, you know, apple of thelate eighties, to the late, went
through a 10 year fallow periodwhere, you know, Microsoft
propped them up right then.
The next team came back in underSteve's, guidance and, know,
they decided to make the switchfrom Power PC to Intel, and they
had a really nice run fromwhatever it was, two th you
know, os 10 OSX,
Rob (07:21):
been a while.
I mean, it's, it's been 25 yearsthat they've been solid.
I mean, the iPod.
It is really what cemented theirposition.
Yeah.
It revolutionized how wechanged, how we listened to
music and the ads they had withthe white headphones and that
killed off Sony.
Basically.
They were like, you'd com andthen the Windows version of
iTunes was a massive deal, andthen ultimately, obviously it
(07:44):
all got killed by the phone,blah, blah, blah.
We've done over that.
But can they keep doing this isthe big question and I don't
think they can, I think they'reout of ideas of where to put new
products.
They've got one everywhere.
I guess there's the some newthings.
There's the, there's the new carplay, which I think we talked
about before and said it was aalso on that missing list that
(08:07):
we had there.
It seems to exist.
Maybe they'll announce some morethings about that.
My big question is, are theygonna talk about the store and
the new roles from the store?
Obviously things have reappearedFortnight's Bag, biggest app on
the uh, app store right now,
PJ (08:23):
will,
Rob (08:24):
I.
PJ (08:24):
be presented as if it was their decision and it was a win.
Dave Lewanda (08:28):
yeah.
Or, or they just won't even talkabout it.
And, you know, I mean, they'redoing it because they've been
basically, you know, consideredper, you know, committed perjury
and, uh.
We're forced to do it.
Um, I think it's gonna be, it'sinteresting, you know, I think
we've even seen, you know,there's been a transition over
the last few years where insteadof crushing everything out in
(08:48):
the.zero release of the OSS inSeptember, that they've even
already started phasing out.
Like there've been some moresignificant, you know, I
remember it was, uh, groupFaceTime was supposed to, you
know, didn't come out to likewhatever, dot two or there, even
on the dev side, there was somechanges to Swift UI that came
out in the dot four release thatwere pretty substantial that you
would've, you know, midyear ormuch bigger shakeup than that.
(09:09):
So I part of this move to likerebranding the OS by the year
that, know, maybe they evenstart, you know, start using the
month somehow in there as well.
And
Rob (09:19):
Maybe they could put it after a dot.
So they could do 26 dot four forApril.
I am a genius.
PJ (09:28):
You have that guy a re.
Rob (09:31):
Exactly.
Uh, Paul, if you're listening,you can't have that idea.
Dave Lewanda (09:35):
It's been trademarked patented.
Rob (09:38):
Yeah.
Dave Lewanda (09:39):
Uh, tricky bits exclusive.
Um, no, I think, I mean, therethere's been some softening,
there's been
Rob (09:44):
I.
Dave Lewanda (09:44):
I think there's, I mean, who knows what's going on.
I mean, supply chain, you know,the, the whole political
tariffs.
You know, I've heard things thatlike, know, China's blocking the
people that Apple want to takeoutta China to go to India to
make new devices.
So they probably have biggerfish to fry of just continuing
to produce hardware.
I don't know even know how theycould produce new hardware if
they're having trouble buildingexisting, you know, the next gen
(10:06):
of whatever the existing productlines are.
So, that further cements leaninginto the services and the
software products that don'trequire manufacturing or import
export limitations.
There was some mention I readtoday around, uh, you know,
again, more playing on theedges, like the gestures that
you can do with your AirPods,you know, where you can shake it
off, you know, shake your headno to tell it to stop or, there
(10:27):
was talk about camera controlsperhaps.
so I think they've made somelittle inroads again, as you
guys mentioned, where these,it's not trying to be that super
agent or agentic ai that can beyour Jarvis or, uh, you know,
complete control.
But it's these little thingslike I've noticed when I am, you
know, getting in the car, itsuggests, do you want directions
to home?
know, those are the types ofthings, you know, it's
(10:48):
recognizing the pattern that,oh, you know, we go down to see
my in-laws every weekend andthat, you know, it's Sunday
night, you're probably drivinghome from where, you know, this
place 45 minutes away.
So can they continue?
But that doesn't make for a goodstory that they can sell
probably until maybe they canbuild it back up.
And that's why I think, youknow, given this year they'll,
you know, jazz hands anddistract away from it.
(11:08):
And then next year they'll rollout all these little
improvements up into a, a betterstory.
Once they have a better story.
But you can't, you know, whatthey tried last year was lead
with the big news and then fillin the details.
Um, I think now they'll probablytry and get, grab the, you know,
gather the details before theyrelaunch the bigger story.
PJ (11:23):
Let me ask you guys this one, o obviously, you know,
there's a question of what wethink WW CDC is gonna have.
What would you guys want to haveit to have to make it
worthwhile?
Like what would it be like, Hey,I'm really excited if Apple
would solve this particularproblem.
(11:45):
Like, what would be exciting foryou guys?
Dave Lewanda (11:47):
Where WW DC has lost some of its shine for me is
that it's not really as much ofa developer's conference
anymore.
day really feels more like aproduct marketing.
You know, dog and pony show andyeah, they've got the labs or
the, the videos that you canlearn stuff from, but somehow
getting back in touch.
And, you know, I've heard a lotof, you know, even some of the
bigger name developers or otherpeople in the Apple ecosystem, I
(12:07):
mean, John Gruber tore him a newone regarding, you know, they're
not even gonna have a allow anApple person to speak to Gruber
during his, you know, livepodcast that he does at WWC for
the first time in 10 years.
Marco Armand and a couple ofthese other big Apple platform
developers are kinda like donewith Apple.
Xcode has become a bear, youknow, to use,, whether it's, you
(12:28):
know, the macros in Swift, know,having to be compiled and, you
know, running up your buildtimes.
They just released apre-compiled version of the Ma
Swift syntax that is requiredfor macros.
But it, everyone I've read is,it's been tough to integrate
Swift UI previews fails everyother time for me.
And maybe it's part of the, toyour point, Rob, around the
store of like getting back,recognizing the third party
(12:49):
developers that built the Appstore or the App store was built
on the backs of all thosedevelopers and the profits from
the app store.
At least getting back in touchto the developers and really
deciding are they going to beyou know, work with developers
even up to Epic Games
Rob (13:03):
No,
Dave Lewanda (13:04):
going, or is they just, otherwise they might as
well just close the system andmake everything web apps like,
you know, the
Rob (13:09):
I think they'd like that.
It's Apple have never likeddevelopers.
They've always been a thorn inthe side because we've always
said like Apple liked to be thesmartest people in the room.
And when you bring in developersor experts in their field,
they're not the smartest peoplein the room.
This happens a lot in games andgraphics and things like that,
and it's Apple have never givena shit about their developers at
(13:31):
all.
It's like, this is what you'regoing to get and you'll like it.
And if they cared aboutdevelopers, they would've never
have made swift.
Why'd you force on language onit just for you?
It's like, yeah, it's opensource, but no one's really
going to use it.
'cause it has no benefit that ifI'm gonna learn a new language
today, I'm gonna learn rust notswift, but generally I'm just
gonna stick in the c plus plusworld because that's what apps
(13:52):
are tend to be, at leastplatform apps tend to be written
in.
And if you're not gonna doplatform apps, you may as well
do web apps, in which case youhave the whole ecosystem of web
apps for development.
So I don't think Apple Careabout'em from my point of view
in the graphics world, if they'dmake a native version of vol,
it'd be excellent.
make it C.
So we don't have to useobjective C to call any graphics
(14:14):
API.
Which again, if they cared aboutdevelopers, if they cared about
games as they say they do, theyjust bought a tiny game studio,
then you'd have an interfacethat.
They want not telling them whatthey want.
It's like they already haveplatforms they can publish on.
They don't need the Appleplatform.
If they really wanted them,they'd accommodate them rather
than telling them what they'regonna do.
(14:34):
But Apple, I don't think arecapable of accommodating
anybody.
It's, this is we all thesmartest people in the room.
And that will trip'em up in theend.
PJ (14:41):
These are actually really interesting answers.
'cause I think this is a greatdouble down on the DC in ww.
DC stands for Developerconference it's fascinating
because I, I agree with Dave's,notion that, you know, the
keynote has become basically a,a product dog and pony show
Rob (15:02):
but before you go any further, that's true for GDC
too.
These days it seems like alldevelopment conferences have
become whizzbang, flashy thingsand don't, don't have any actual
content.
GDC still has some content, butit's getting down that same
path.
WW DC, the Microsoft one and theGoogle one and the Facebook one,
and they were all just flashymarketing events of, look, we
(15:25):
have developers.
This the days of just a backroom of developers sitting in a
room and hashing it out arepretty much gone.
And that may be because ofYouTube and the way we consume
media today.
It's, and all the informationyou want you can find online.
They don't really need to sityou there.
And secondly, if you go into ameeting with these people, you
(15:45):
can get their info and theydon't want your content in them
directly.
They want you to go throughproper channels, blah, blah,
blah.
There's many reasons why thedeveloper conference has kind of
faded and it's become thiswhizzbang thing, but I think ww
DC is the worst offender.
Maybe Google IO is actuallyworse, but it, it's close.
PJ (16:04):
bad up there.
Yeah.
So, so let me then pivot thequestion.
Do you guys feel, and maybe theanswer's an easy yes.
That these conferences nowoccupy the space as conferences,
really for the shareholders andthe news media and the product
people that want to see the newwhizzbang features, rather than
(16:25):
being about how do we empowerdevelopers to operate better on
our platforms.
Rob (16:31):
Yes, it is exactly that.
I think, and you can see thisbecause it's always from their
point of view, if like Applewill tell you this is how the
new API will be used.
But if you go to like a realdeveloper conference, like I
know the PlayStation's GTC oruh, the internal one for first
party.
Studios, they'll get developersup there saying, this is what we
did.
(16:51):
We told it wouldn't work.
We did it this way.
These are the issues we had.
This is how we worked around it.
The end result is this andthings like, this is how we got
Nan night and visibility buffersand all these new technology.
Was someone just going, what ifI did this?
And it's like, oh crap, it worksreally well.
You don't get that opportunityon Apple platforms.
(17:11):
I will still, I'll go on a stageand say Microsoft's developer
support to this day is athousand times better than
Apple's and'cause they don'treally care about their
developers.
Where Microsoft document all theAPIs, they are very good at like
the MSDN old days and then MSDNonline now it's, it's orders of
magnitude better.
(17:32):
Apple is just literally, they'lldocument the function calls and
it's like, I can read that fromthe damn header file.
Dave Lewanda (17:38):
Yeah, well that's, that's one of the funny things
is some of the documentation,like maybe that would be better
if they just canceled thedeveloper's conference and put
all the money and resources intothe documentation.
'cause often you'll find stuffthat's not in the documentation,
that's in a video, like a ww DCvideo that gets published, you
know, some Edge Corner case forsomething I found, you know, uh,
how you control the center stagecamera on an iPad was not at all
(18:00):
documented in the, traditionaldeveloper documentation, but the
guy who did the presentationmentioned it and, and had it on
one slide in a developer video.
So, yeah, I mean, I don't knowif there's, I.
There's a way, you know, if youwanna have a dog and pony show
for the product team, just do itand call it that launch the iPad
in, June, if you're gonna do thephone and the watch in, uh,
(18:21):
September and the computer, youknow, the Mac in November.
yeah, it really has felt like,it just doesn't, it doesn't line
up with filling the needs of thedevelopers necessarily, but
maybe that's it.
'cause Apple, I mean, that's, toyour point, Rob, I mean,
Microsoft has always been aboutselling their software and
they'd be, would've never madeit anywhere without the third
party developers.
Whereas Apple just wanted tosell you hardware and now it's
hardware and their services.
If, you know, you sign up forApple Fitness and Apple TV and
(18:42):
Apple News and Apple iCloud Plusor whatever and never use a
third party app, they'd be,they'd be fine with it.
Right.
Rob (18:49):
going back to your question pj, what would I like to see?
I'd like to see Apple open uptheir platform of like, these
are all the internal APIs thatwe use.
You can use'em to, that ain'tgonna happen.
And I think if that did happen,you'd see how insecure the Apple
platforms were.
Uh, which is kind of why theykeep doing these wall gardens,
even as the.
And Microsoft does that to someextent, but they are pretty good
(19:10):
at documenting or at leastacknowledging that these are
there.
They say, they don't say youcan't run them, they just say
you probably shouldn't.
and again, good things have comefrom developers doing that of
like, oh, you said don't callit, but we have to,'cause we
have to do A, B, C, whatever ourproduct requirement is.
The requirements of your productis more important than Apple
saying you can or can't do this.
(19:30):
But on.
OSX.
Sometimes you just can't do it.
It's like you absolutely can't,especially if your app's gonna
go in the app store.
'cause then they will vet thatyou're not calling any of these
internal things and you have tonow go via alternate
distribution paths, which forMac you can still do, but iPhone
you cannot do.
So that's what I'd like to say.
Open up and just be more honestas to how these things are
(19:51):
implemented.
And you're not the smallestpeople in the room.
Developers can help you.
They will help you.
They'll tell you better ways ofusing these APIs, better ways of
structuring them.
But you have to be willing tolisten.
And I'd also like to see sideloading on the iPhone.
Get rid of the$99 developer fee,which if you could side load,
you have to get rid of that feebecause why?
(20:13):
Why would anybody pay?
So there's all lots of thingsthat fit together, which just
everything that sucks aboutbeing a developer on the Apple
platform, I think we just talkedabout.
PJ (20:21):
To round back to the same question you, Dave, what are the
things, what are the tech.
Technical things.
I mean, you talked aboutdocumentation, but what are the
technical things you would loveto have coming out of Apple that
they would announce at W-D-W-D-Cthat you'd be like, yes, that is
awesome.
That is worth it.
Dave Lewanda (20:39):
I think just modernizing their APIs., The
mixture of, callbacks and, thereactive approach was hot for a
couple years and then, you know,now you know, async weight, you
know, mean swift.
I liked Swift.
I got into Swift.
I, I thought I had a lot ofSyntactical cleanup, you know,
more concise, you know, andeasier to read, easier to teach.
And I think that's kinda wherethey were going.
But it up through Swift five,like Swift 5.9.
(21:02):
It of a couple years ago, andthen they started to move
towards Swift six, which is thebig move to the concurrency
stuff.
It all started to fall apart andit's become a, a mess.
And then the fact that like alot of the Apple APIs,
especially the Apple frameworks,are just a mix.
They, they haven't bothered, youknow, the, uh, keeping them at
all in sync.
You know, each one is slightlydifferent, different generations
of callbacks or key valueobservation or having a
(21:24):
publisher interface and nothaving any level of consistency
there.
So even if you wanna structureyour app, you end up having to
wrap it, your, each one, eachone of their SDKs, whether it's
core Bluetooth, whether it's AVfoundation in a way that works,
to build your own.
Consistency on top of that,which makes, you know, the
mobile app of any, any level ofcomplexity that much more
complicated.
(21:45):
'cause all those things are sortof, can move underneath the hood
for you.
As we, as I dealt with in, youknow, jobs.
There's gotta be an answersomewhere here around, in the
blog post I wrote up aroundhaving a single platform for
these mobile devices, you know,I don't think it's React native.
I, I'm not sure it's flutter.
There's been a big push for,there's a Swift on Android group
and maybe they'll talk if theytalk about that at wwc.
(22:05):
Um, apparently they just did abig.
post, I haven't had a chance toread it yet on how they use
Swift to rewrite the swift, onthe server side for the
monitoring for the ApplePasswords app.
the Mac, you know, the AppleiCloud passwords app that they
pulled out of the OS and made astandalone app, probably because
of competitive, concerns, withthird party vendors like One
Password or LastPass., So Ithink there's opportunities
(22:27):
when, when Swift is used, well,it feels like it can be useful
and, you know, can be veryexpressive.
I've, I've enjoyed that, but italso can get very messy and,
know, just be a, a dog on a lotof these things because
especially you know, theconcurrency problem, they try to
solve a lot of the reallychallenging stuff, but messed up
the easy part.
They pushed everybody into thedeep end of the pool, but before
(22:48):
everybody was ready to, and nowthey're that, and Swift 6.1, 6.2
is tried, but I haven't, and Ihaven't been as close to the
code in the last few monthsmyself, but everything I've been
following online suggests thatit's, you know, they've, it gets
messier before it's gonna get
Rob (23:01):
Isn't that a symptom of being the smartest person in the
room?
Dave Lewanda (23:05):
Well part of it, the, the guy Chris, uh, Latner,
I think was his name, I'm mightbe mispronouncing that, who, was
the, one of the originalfounders of Swift Left Apple.
And I think that's sort of when,you know, he's the one who put
together the originalconcurrency pitch.
similar now, to, Steve Dobbs,when who take over for the
vision lose sight of the, thereasons and the, the motivation,
you know, they're trying andthey're probably not, you know,
(23:26):
I don't, again, don't wanna bessmart to anybody, a lot of smart
people who are doing it, but,you know, maybe they're
overthinking it.
I don't, I don't know.
It's a, it's clearly a hardproblem to solve,
Rob (23:34):
It is, but they never do it in like a layered approach.
Like you could just be like, ifyou took the C approach, even
let's just take threads in theoperating system as we know'em
today, they haven't changed in50 years what they are.
That's a thread.
If you want to do asynchronousat some point you create one of
these and, and then on top ofthat, you build various.
(23:54):
Asynchronous models and whohandles creating and destroying
the thread, who handlessynchronizing them, who handles,
when they run and when theydon't run.
Things like that, that can allbe in the higher level
frameworks.
And even Swift, it's in its highlevelness, it's still doing this
at the operating system level.
So I think the whole vision oflike, any of these layers you
(24:15):
can interject at based on whatyou want for your app.
Apple never do that.
I, I'd love to see them becomemore of a, a straight up.
Like this is the platform, theseare the levels and layers that
we have.
And comes back to what I saidabout Volcan Avenue C interface.
it'll never happen, but it'd benice to do that.
And I still think a lot of theold fashioned platforms like WIN
(24:36):
32 or X 11 or things like thatare still have that thing.
You can use these big fancyframeworks on top.
But generally it's built on thisand it's that this bit that's
missing on the Apple platforms.
And I've always thought thathaving to throw in a new
language and a new model and anew concurrency model and all
that on top of everything elseis just, I think it's hard.
(24:57):
I think Apple's a very hardplatform to develop on.
Xcode you mentioned is horrible.
It can't even give me errors ina file in order it has to order
them.
For me, it's like, why can't youcompile like every fucking other
platform compiles of justcompile a file, tell me what's
wrong with it.
Dave Lewanda (25:12):
best one.
If, and not that you needed anyre convincing that swifty y is
bad, but right now it'll justsay it can't compile in a
reasonable amount of time andyou end up having to do a binary
search by like commenting outthe whole block and then
uncommon half of it, see if itcompiles, and if it tells you
what the real error is.
Nope.
Comment that back out uncommon,like that feels like a just
complete miss.
As far as the tooling on thetooling side,
Rob (25:32):
Yeah.
And annoying things like thatwindow with the errors in, it'll
scroll up.
there's so many inated areaslike show if you know this one
caused it, pull it at the bottomso I, or put it wherever the
focus of the window's gonna be.
it's like, don't make me scrollall the way back up and do it,
at least to the CI mean, ithappens to see too, but at least
it's like file by file.
Like, I know these are all goodand this one isn't.
And it's just ridiculous thatthey do things so different to
(25:54):
everybody else where there's noneed to
Dave Lewanda (25:56):
I mean, I think going back to your question,
like what would I want if, if Icould have one wish, you know,
the genie came outta the bottleand I had one wish for the Apple
Developer Conference.
Fix X code, just make X code
Rob (26:08):
make Xcode Visual Studio.
'cause that's the, that's theone thing that Microsoft's
always made.
Well, I was, I've never been ahuge, huge fan of it.
Mostly because it has a terribledebugger, but compared to the
debugger in X code.
Visual studio might as well besome like kale embedded
debugger.
It's, it's great.
Dave Lewanda (26:27):
I'll, tell you still my favorite ID of all
time.
Uh, and even with all the workthat's gone into like the AI
enabled stuff not that I'vespent a ton of time with, that
is still net beams that sun putout for Java swing development
in 2006.
It was amazing.
It would, you know, couldrefactor code every, every which
way you could almost to thepoint, like similar to what you
(26:48):
see some of the LLM stuff doingnow.
It's like, I need a function.
It would fill in, it wouldautomatically fill in blocks
and, you know, help you, youknow, complete the code in a way
and point out where the bugs,the errors were.
I mean, you'd never have acompiler.
It almost had, it felt like aspell checker, like as you were
typing it would, you know, getyour syntax and tell you the
wrong thing.
And
Rob (27:05):
See, for me it's not about the idea, it's about the
debugger.
It's like if you went in lowlevel code or very technical
code, I just wanna step throughthe function and see what it's
doing.
And I think the debugger is thekey.
And for me, SN Debugger for thePlayStation was brilliant.
And then Sony killed that'causethey integrated everything into
Visual Studio and it's gettinggood again now.
(27:27):
But it's took them well 15 yearssince the 20 years since they
killed the SND bugger.
Uh, I think the KLD bugger, thearmed debugger is really good.
Sega Ozone Debugger also anembedded debugger.
Really good.
Uh, mostly'cause a lot of theseare scriptable debuggers.
So you can write scripts to dothings that don't have function
(27:49):
native functionality within thedebugger.
Technically excode can do it'cause you can do python in
A-L-L-D-B, but uh, it's soobfuscated that it's an exercise
in frustration just to get toit.
Excos always been the worst ofall the ideas for sure.
but it's interesting'cause allthe things we're talking about
should be things that are at adeveloper conference.
(28:09):
Enough.
Nothing that we just talkedabout for the last hour will
even be relevant at ww DC'causeof the, the whizzbang and the
flashes that they need to putout there, as PJ said to, to
keep the shareholders happy morethan anything.
And we are doing something, weare making progress.
Dave Lewanda (28:24):
not the the old crusty guys like us, they're
trying to get, you know, newpeople, you know, a more diverse
developer crew that a lot of thestuff is, you know, geared
towards the student developer orthe, you know, making it as easy
as possible, lowering the, theon-ramp.
A lot of, you know, the swiftUey stuff was clearly to move in
that
Rob (28:41):
Well, if you go in that direction, do what you said and
do a web app and get it for allplatforms at the same time.
Dave Lewanda (28:46):
Again, they're trying to lower the barrier to
entry to get people, you know,so that, you know, those kids in
20 years are doing a podcastabout how bad Apple is in 2045.
But, who knows what they'rereally thinking?
It's clearly a hard problem.
And, you know, again, I'm nothere to, you know, drag him on
anything, but he certainly coulddo better.
It feels like, it doesn't seemlike it'd be that hard to do
better.
Rob (29:03):
Oh, I'll drag over the cold.
not a problem.
Doing that, having worked there,I.
Internally, it's as messy as youthink it is of like I can tell
you exactly how those frameworkscome become to be so disjoint
and all of that.
And they do have an APIstandards committee inside.
So you have every API you haveto submit to it.
And they do supposedly to dolike style of API auditing and
(29:27):
they also do security auditingand things like this function
could be a problem, but theydon't, they're not by good
because loads of crap gets pastthem.
They have standards like inteshave to be intes.
Like if it's, even if it's anunsigned value, it still has to
be an integer because in 64 B,you're not gonna have a number
that's bigger than you can storein 64 B, but, but things like
that, it's like you're being thesmartest person in the room when
(29:49):
you enforce things like that.
And I've said this exactly asI'm gonna say it again on this
podcast before is.
Unsigned INT is a safer datatype because if you pass in a
negative number, it shows up asa massive positive and the check
you have to say, if X is lessgreater than 10, we'll catch all
negative numbers and only zeroto 10 will get past it.
(30:12):
If you are suddenly forced tohave that value being integer,
because some API standards boardsaid, so you now have to
remember to go back andexplicitly check for all the
negative values and be like, ifX is less than zero, then it's
also an error.
And lots of times and lots ofbuffer overruns and lots of
security problems have comeabout because you can index
(30:33):
negative in this functionbecause no one checked the
negative value.
It's just things like that of,and even in the on go into a
whole different subject on theVision Pro, there was swift
inside system code and there wasa, a system wide, architecture
decision.
Now you can't allocate memory.
Which basically means systemcode has to be C, which is what
(30:53):
it should be, not even c plusplus in the grand scheme of
things, it should be C, the OSis straight C.
The system code should bestraight C as well because you
can control when it allocatesmemory, objective C and Swift
will even objective C will dothis and Swift does it more, but
objective C does it.
When it reallocate its dispatchtables'cause it's like, oh I
need to reallocate the cash bangmemory allocation.
(31:15):
we didn't prevent memoryallocation by saying you can't
call Malo.
We blocked the call.
So to make it even worse is Malowould sometimes work'cause you
go, I need a K and it's got a Kin system in a user space, so
well give it to you and it's allgood.
But then every now and thenMalik's like, well I need
another page from the operatingsystem.
So it'll go and hit the SIScall, then it will crash and it
(31:35):
crashes miles and miles awayfrom the thing that actually
caused the initial problem.
And that's a good thing reallyyou do want, thats call to be
blocked.
But if you're gonna do that,you've gotta structure your code
above it in a way that'sexpecting, thats call to be
blocked and know what things aredoing.
People would just show up and,oh, I'll just write this in
(31:55):
objective C And I've said thisbefore too, we, that API that
would pass a dictionary inbecause people couldn't agree on
what the fucking parameters forthe function code should be.
So we'll pass a dictionary in'cause then you can just query
it and see if your parameter'sthere.
And if it is, you can do this.
And if it isn't, you can read adifferent parameter and all that
of like, how bad do you wantthis to be?
(32:16):
And it's system level code thatjust gets a billion times worse
when you get up to the highlevel code.
They do not have the rightpeople in the right places is
really what it comes down to.
And we've had a podcast on thistoo.
It's.
People get hired and they getmoved around.
And even on my team, there was aguy who was brilliant at math
(32:38):
and all this and he was doingembedded code and I was doing I
level code and we could have alljust took his position to the
left and would've all been in amuch better position to, to do
things.
But not because the head countand how the teams work and all
of this and who needs good staffand who doesn't it, it was just
a typical middle managercorporate mess.
PJ (32:57):
Welcome to Big Tech Folks.
Rob (32:59):
It's a mess, but they, they have this API board to try and
fix it, and they enforce thingslike ins have to be ins and not
uin, and it's like you're notactually fixing anything.
You really have to go back towhat we said about GDC and open
up the layers, clean it all upfrom the ground up, and not
build all this on a basic houseof cards.
(33:20):
It needs to be solid foundation.
Then you can go to the highlevel languages on top of that,
but the foundation has to besolid.
But most people have thatlayering system Windows as if
Linux, as if Apple have it atthe OS level, like Kernel is
clean and fairly good.
Above that, it's fucking freefor all.
Dave Lewanda (33:40):
I wonder if there's a space for, you know,
if that's where AI could beuseful is, you know, turning
what people are asking for intowhat they, the way they should
do it and you know.
Things, from running amuck it'smore regimented and it, you
know, could learn over timemore.
So, you know, like self-driving.
Could you have self programmingwhere you say, I want to get
from, you know, instead of Iwanna go from point A to point
B, I wanna get this piece offunctionality and then have the
(34:03):
AI really enforce more of thestricter rules, more rigidly
than, you know, humans are.
'cause humans will always fightover, you know, the bike
shedding of, you know, un versusin or four, you know, four
spaces versus a tab versus twospaces
Rob (34:18):
Two spaces is the answer.
There's no argument.
It's two spaces.
PJ (34:24):
All right.
That'll be a topic for anotherpodcast.
So for ww DC we would love it tobe about the DC won't be.
It sounds like DC will not standfor de developer conference.
It'll stand for damage controlfor, for AI at, at best, and
maybe a few passing featureshere or there.
(34:45):
Gentlemen, Are there any majorfeatures we expect to see out of
this next conference, even froma product shareholder marketing
standpoint, or is it gonna be ofsmoke and mirrors around minor
stuff like we've seen in thepast?
Rob (35:03):
I don't know.
I, uh, I'd like to see some newhardware, but I don't think even
the rumor mill doesn't reallyhave much new hardware.
A lot of it's been announcedjust along the way, various
releases here and there.
Maybe they're stepping away fromthe, the all in one hardware
announcement.
Maybe there's a new Mac studio,an M four based one, maybe the M
four ultra, whatever the JUULone is.
(35:25):
I think that's the ultra, maybe,I don't know.
But I think if there's anyhardware, it'll be fairly small
things for a refresh.
Dave Lewanda (35:33):
M five chip, you know, because the M four has
been out for, you know,announced that the M five is
coming later this year.
AirPod Pros, maybe I could see,you know, the, now that the
AirPod pros are like, you know,medical equipment and that
they're hearing aidreplacements, they haven't
really touched that.
I would love to see the home podwith a screen, but, you know,
that, or even just better Homepod support, that that's been an
(35:55):
area that they've let languish,you know, I was hoping it'd be
kind of like the Apple tv, youknow, the original Poppy project
as Jobs called it.
And then they turned it intosomething a bit more real over
time that like the home podfeels like it's ready for,
coming to the table as a real,as a real player, again, putting
a screen on.
I, I have a perfect use case inmy fa in my, you know, kitchen.
You know, if I had a screen thatI could put like a family
(36:16):
calendar on you, talk to the,you know, with, especially with
Google stepping away from theirhardware.
More of the home kit space.
Some more support for just smarthome.
But I dunno, that feels likeit's fizzled
Rob (36:25):
I think.
I think that's the solarhardware you're gonna see.
You're gonna see the smallerthings.
There's no new phone.
There's no new I.
There might be a new iPad.
I don't think so.
Maybe a new web of one of them.
I think they've
Dave Lewanda (36:35):
I think, you know,
Rob (36:36):
done them all though.
Dave Lewanda (36:36):
studio display, maybe the next generation studio
Rob (36:39):
A big iMac.
I would like if they'd make a 30inch iMac, I would buy one.
Dave Lewanda (36:44):
that all that's gonna be limited by, you know,
with the political and tariffsand import export challenges.
I, I think maybe that's wherehardware just is DDOA this year.
Rob (36:53):
Maybe.
PJ (36:53):
make a prediction and I, I suspect this is something they
will use the, tariffs as a coverstory for, I think they will
deprecate the Mac Pro.
I think it has existed basicallyfar beyond its useful lifetime
they can use the cover story ofthe political stuff as
effectively a way to get rid ofit.
Rob (37:17):
I think we had a whole podcast on this exact thing as
to hard take on the Mac Pro andI would, I think they announced
they, we said this before I.
They released it because theysaid they would, it's an
abomination of what theyactually released.
You can't put new memory in it.
You, you can put PCI cards init, but you can't put a GPU in
it, blah, blah, blah, blah,blah.
(37:38):
And I could see them just beinglike, yeah, it's, they did it.
They said they do it and that'swhy they did it.
And now it's time to let itmaybe die quietly.
PJ (37:47):
Well,
Rob (37:47):
And
PJ (37:48):
can use, they can use all the political
Rob (37:51):
I think so yeah, it's time to let it die.
It's like, let it die quietly.
And it's.
PJ (37:55):
a new Max studio like you guys are saying at the same
time.
To say this is occupying thishigh-end space now.
Dave Lewanda (38:02):
Yeah, only'cause I've heard, you know, with what
and seen the stuff like maybemeta's done with glasses.
Could they have another visionPro or vision not pro their
sleeve somewhere?
Rob (38:15):
They would, they'd be more leaky things about it if it
happens.
So they were looking at thatsort of thing when I was there
and there is a second visionprobe being made, blah, blah,
blah.
but I don't think we're in amarket space that wants that
product.
and I think meta may be gettingthis more right than Apple is
(38:36):
when you have a new interfaceand you alerted this earlier
with the keyboard mouse going tothe touchscreen and go into the
ai uh, products.
If you have glasses that are onyour head and they're gonna be
there for a long time, you'vegot to ease people into it.
You can't just be like, oh look,here's a full experience.
And I think if we had glassstyle, like meta style glasses
(39:00):
today that were full ARexperiences, people wouldn't use
it.
But.
If you take a small step and be,okay, I'll just put your
notifications in your face soyou can look at them while
you're driving.
I'll pull a camera in it so youcan take, video.
Basically things that meta did,that's the stepping stone that
people need to get used tohaving this thing on their head
(39:23):
all day.
You can't put this thing on yourhead and have this intense
experience and expect to wearthese for 12 hours.
Dave Lewanda (39:29):
That's where I, think maybe again, the AirPods
have a space there to be thatstart being that bridge.
You could, you know, put alittle camera in the, in the
stem somewhere, or eveninfrared, you know, some level
of sensor, you know, if they'redoing, getting you to do
motions, like shake your headno.
Or, you know, those type ofthings.
You know, more, more smarts inthe
Rob (39:46):
yep.
Yeah.
Dave Lewanda (39:46):
again, way not to cover your face, you know, I
mean, a number of folks havepointed out, you know, just
having something on your face asa detractor, know, losing your
peripheral vision and or, youknow, your ability to track
mates a, uh, thing that, youknow, just human evolution has
shied away from.
So, you know, can you do it allfrom your ears?
And maybe it's, you know, biggerear pods or bigger, you know, a
(40:07):
headband that doesn't go overyour face is there, you know,
other places to put hardware,whether it's tying to your phone
in your pocket to your watch onyour wrist, to your, I don't
know what other, where else youcould stick, compute or, you
know, bandwidth that would work.
But, Maybe just not gettingaway, getting away from the
face.
And that's where I think maybethe AirPods have some space to
grow and, and even especiallywith them till off the cans.
(40:30):
'cause they realize the cansaren't a solution, but more
focusing more on the, theearbuds.
Rob (40:36):
I think we missed something else too.
There's some game stuff thatthey're gonna announce.
They have that new games Xboxstyle game center that they're
gonna announce.
Dave Lewanda (40:45):
app,
Rob (40:45):
just replacement for Apple Arcade.
I think it's just, I think abetter way to find them, but I
think it's still, they're notgonna take games seriously.
Dave Lewanda (40:53):
you know, they pull it out of the app store and
make it into a, like, you know,apple News for
Rob (40:58):
Yeah.
Dave Lewanda (40:58):
and Apple Sports.
Rob (40:59):
it'll be.
Yeah.
And I still think, and I'vealways said this, they're
missing a huge opportunitybecause the Apple TV is quite
powerful.
It could do some pretty decentgames if they opened it up
console style to actual consoledevelopers.
Dave Lewanda (41:15):
and they wanted to, with that original remote,
you know, that had the motion,you know, you used to have like
a Wii style remote in that, theone that had, you know, just the
flat
Rob (41:23):
Yeah.
I mean, it'll connect to aPlayStation controller too, of
like, if, I dunno what that saysit though, is that Apple saying,
we can't make a bettercontroller, so we'll just use
yours?
Or is that Apple saying, wedon't give a shit, we're not
gonna put effort into it.
We'll, uh, we'll just use yours.
PJ (41:40):
My suspicion basically is that they won't get into the
game space because of the, thedomino effect, Rob, that you're,
you're kind of alluding to isthat really do it right, you
would need to put Vulcan assomething native on all of these
to do that.
You then say, okay, well thenhow do I open up the OSS in such
(42:04):
a way that everyone can seeeverything?
In which case I think you hitthe security issues.
So I, I suspect that there'sthis cascade of stuff where it's
like, if you really want to dogames right on the Apple tv, on
iOS, on the Mac.
You end up actually opening upthis gigantic can of worms that
Apple simply doesn't
Rob (42:25):
And it's an ecosystem that already exists here.
It's if you use, if you're gonnamake AAA game, you're gonna use
Ws and you're gonna let it dothe mixing, and you're gonna,
you need basically access tocontrols, video and audio.
Network sockets.
And outside of that, everythingelse could be layers.
But now these layers thatalready exist don't fit this new
(42:47):
framework.
So we're saying, I think it maybe part of what you're saying,
Dave Lewanda (42:50):
if not anything, it feels like they tried, they
did that when they did that, youknow, scary event.
It was Halloween a couple yearsago, and it was at like 8:00 PM
Eastern, so it could be on andduring the daytime in Tokyo.
And they had that famousJapanese game designer
Rob (43:03):
they always have a game.
Every single time they release anew GPU, they always have AAA
game, but they've paid them toport that and they jump through
the hoops of calling theobjectives and all that.
So I'm not saying you can't doit, I'm just saying developers
won't do it because it's the oddball.
Dave Lewanda (43:20):
Right, because well, yeah, so I mean, it feels
like if apple apple's dippedtheir toes in all those times
and never done it, they're,they're never going to tip pj.
To your point, they're nevergoing to do it.
PJ (43:29):
I, I just think it takes way more than they're willing to
commit.
you have to start from theperspective that I care about my
developers, in which case youhave to ask your developers,
what do you need?
Rob (43:42):
Yeah, they're capable of doing that.
PJ (43:45):
that step, then again, you can pay people to jump through
your hoops perfectly willing todo that, but you're not gonna
get the, the best, uh, the bestdevelopment experience, the
Rob (43:59):
Games is the extreme example of you can't do it
yourself I mean, so is kind ofgetting there by buying all the
studios, but the fact they ownthem doesn't mean that they have
any say over them.
think PlayStation by itself isuseless.
And we've talked about this likePlayStation will win every
single console war because ofits exclusive content, which
(44:20):
come from developers.
Like, so don't make thatthemselves.
Like I says, they are graduallybuying up all the developers, so
technically do own them, but,uh, it's still independent
developers, independent mindsetof developers at least that
makes these really awesomegames.
And Dave said it earlier ofit's, these are the frameworks
(44:40):
you get and do use it.
And if, if they can give yousome services and give you
something that you'll use as is,then it's, it's what they'll do.
And again, it's the smartestperson in the room syndrome.
It's, it doesn't work for gamedevelopers and there's never
been a platform more extremethan PlayStation for, without
the developers, this platformwon't even exist.
(45:04):
Unless you're going to, I thinkthere's also a gray area, like
what is, we talked about thistoo.
What is the future of consoles?
Maybe it is Apple tv.
Maybe it is like get rid of thebig consoles and just have
something that could stream fromPlayStation Cloud or Xbox Live,
or whatever it is.
So maybe it's not worth goingdown the path, but they already
(45:24):
have the hardware, so it's notlike they're making new hardware
and having to spend a billiondollars promoting it.
They could literally just changea few things and they'd
immediately be a, a verycompetitive gaming platform.
PJ (45:41):
Well, we've got some fun predictions for what Monday may
bring.
I don't think any of us areespecially optimistic at this
point in time, but what we'll dois next episode, we'll do some
grading and some commentaryafter the fact to see exactly
what we got out of this WWDC.
So stay tuned, folks.
(46:02):
This story is not over yet.
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com