<drez>
you'll also need non's tweaked fltk, called ntk
<drez>
I think it's included in the 'non' repository
<riteo>
yeah the linuxaudio mirror already has the submodule handled
<drez>
I think I have Stazed's fork installed
<riteo>
that's a pity though that it went like this
<drez>
it really is
<riteo>
its goals were very, very, nice
<drez>
well, it was developed for many years
<riteo>
not that KISS and creative tools really go hand-to-hand (for various reasons)
<drez>
it's not like it's super incomplete
<drez>
it's very usable
<drez>
riteo, why don't they go together ?
<riteo>
drez: I guess the thirst for every feature possible
<riteo>
and I mean, that kinda makes sense
<riteo>
creative tools usually want to give you every tool possible because it might be genuinely useful for that niche use case that you might need to finish your work
<riteo>
I mean, it's not like blender is that bloated AFAIK
<riteo>
but stuff like Krita is still pretty heavy
<riteo>
or perhaps I'm misunderstanding feature-completeness with bloatedness and un-KISSy software?
<riteo>
I guess that in some cases something might have a shitton of features and still be "simple"
<riteo>
I'd definitely consider Godot "simple" yet it's surprisingly close to feature-parity with Unity for the average indie developer IMO
<drez>
no well
<drez>
you could just modularize stuff
<drez>
but still have all the features
<drez>
which is what non was doing
<riteo>
right
<riteo>
well I guess that's definitely the ideal, despite being the harder choice
<drez>
pretty hard for sure
<riteo>
a bit like the issue with microkernels, they were initially slower and thus got put to the side
<riteo>
although the new ones are blazing fast
<riteo>
only took 40 years :P
<drez>
:D
<drez>
which ones exactly?
<drez>
l4?
<riteo>
yeah
<riteo>
I'm pretty ignorant about the actual history but the main comparison I know of is mach and l4
<riteo>
the first one was slow enough to be actually considered as one of its main con while L4 only had, IIRC, a 4% overhead compared to a monolithic kernel
<riteo>
not shabby for an IPC :P
<riteo>
Yeah the first sentence of wikipedia says: "The poor performance of first-generation microkernels, such as Mach, led a number of developers to re-examine the entire microkernel concept in the mid-1990s."
<midfavila>
the performance problems with microkernels haven't been solved *to my knowledge*
<midfavila>
it's just the overhead inherent in using IPC instead of a monokernel
<midfavila>
but the increase in commodity hardware has rendered that overhead irrelevant
<midfavila>
s/increase/performance increase/
<midfavila>
i love when i just randomly forget to include words in my sentences
phinxy has quit [Ping timeout: 265 seconds]
<riteo>
mh I can't find numbers for l4
<riteo>
I only found this random uni document talking about mach having an 800 cycle overhead
<riteo>
whatever that means
<midfavila>
800 ticks of the CPU timer
<midfavila>
800 cycles in the mach era would have been a huge deal
<midfavila>
(cycles are measured in "frequency", so when you hear people talking about an Xghz cpu or whatever they're referring to how many ticks of the CPU's clock are happening per second)
<midfavila>
but given the speed of modern CPUs and the whole multithreading thing... it's kind of whatever if you spend 800 cycles on IPC overhead
<riteo>
mh
<riteo>
well, L4's being used also for embedded so I presume that it's good enough
<midfavila>
yeap
<riteo>
and in theory the benefits would have been worth it anyways balancing the pros and cons
<riteo>
obv depending on the use-case, but security-wise a capability-based ukernel AFAIK should be a huge win
<riteo>
there are also a list of "features" which also include elon
rohan has quit [Read error: Connection reset by peer]
<riteo>
and political parties for some goddamn reason
<riteo>
I mean, it would make sense to favour elon's posts
<riteo>
but I'm seriously confused by this thing
<riteo>
yeah ok that was the list
<riteo>
I mean as someone pointed out it's also at the end of the list and could've thus been added very lately
<riteo>
IDK this is seriously baffling me
<riteo>
this looks like too much effort, either they wrote random stuff or IDK
<riteo>
it's not like they have the money to put all that effort for a silly april fools
<riteo>
yeah if this is serious I'm pretty confident that it's just a cost-offsetting method. They talk about contributing and hackerone bounties
<testuser[m]>
Hmm
<riteo>
> We hope to benefit from the collective intelligence and expertise of the global community in helping us identify issues and suggest improvements, ultimately leading to a better Twitter.
<riteo>
yeah they broke af lately and are desperate
<riteo>
that's my bet, they haven't the resources for such an elaborate joke IMO