fijal

This is quite fascinating as it looks like the esteemed professors run into a problem of runaway capitalism for the first time. Guess what? The corporations involved don't really care about your feelings or how pursuing their goals affects your life, your well being, anything at all here. Talking with them is kinda moot, this is full on adversarial as countless people around the world, say the nigerians dealing with oil spills would tell you.

  Location: Cape Town, RSA
  Remote: YES
  Willing to relocate: NO
  Technologies: Python, C, assembler (x86, ARM), C#, unity, C++, JS, kinda anything
  Résumé/CV: -
  Portfolio: PyPy project (https://pypy.org), VR Sketch (https://vrsketch.eu), Building and operating climbing gyms (https://bloc11.co.za)
  Email: fijal at gmail

I'm a curious individual working on a variety of projects. I mostly like intersection of humans and technology, would love to talk to clients, translate that to ideas and what do we actually build. Love prototypes.

Is there a single error bar in the paper? It seems like a guess

Hi Dmitry, Hi Ryan

I love the fact that this is out! I have been the original author of vmprof and I have been working on profilers for quite some time. I'm also one of the people who worked on PyPy. We never managed to launch a SaaS product out of it, but I'm super happy to answer questions about profiling, just in time compilers and all things like that! Hit me here or in private (email in profile)

All who live in silicon Valley of course. You seem to have forgotten that "making world a better place" is a mantra repeated across the entire IT. I presume if Facebook employees actively think that, so can RIAA lawyers

That will likely be the contents of the next blog post

It did: https://en.wikipedia.org/wiki/Trolleybus

If you want to see an advanced Python compiler, have a look at PyPy. While CPython bytecode compiler is "dump", it's as smart as it can be. You can do crazy stuff with speculative replacing of bytecodes with guards and type-specific bytecodes, but then you might as well create a full just in time compiler. It's incredibly hard to create optimizations on bytecode level without changing semantics, which Python rightly stayed away from.

Thanks!

I'm glad to hear your company is also incorporated in Estonia. How well has it been working for you?

Hi

I suggest you don't write down in gray on white "by the way, we will also store your email". Put it somewhere either prominent or maybe outright have and opt in button for that? It's an industry wide practice, but that does not automatically make it ok.

There are two things in play here - one is that pypy is a large py2-only codebase that is unlikely to be ported. Second is that maintaining python 2 both does not cost much and it's used by the majority of our users, despite of what you hear on the internet. So, there is both good reason to keep it and a large amount of work to drop it.

While I agree with you, pypy already runs cffi fast and we are thinking how to run cython fast. That would cover a large subset of using C API I think.

I live in Cape Town where I can do surfing a bit better than in Munich ;-)

I think you are confusing CFFI (which is massively faster under pypy) with CPython C API calls (which is both slower under pypy and ones used by numpy)

The main way big corpos are sponsoring os is hiring the developers typically. That comes with a whole lot of things that some people don't want to deal with, like changing your place of residence

3.7 (or maybe even 3.8) support will start as soon as we can have a "final" 3.6 - which might be couple months from now. I think 3.7 won't be done in 2019, unless some magic source of funding pops up.

numpy works on pypy these days. You have to install it via pip and it does not have wheels, so it takes a few minutes to install, otherwise works. The numpy-heavy workloads will not be sped up by pypy, but depending on what you are exactly doing your python program might be faster in other parts.

We are actively trying to secure funding and goals to make numpy fast, now that it works and we are working together with numpy community to make sure it stays working.

And the website reporting it blatantly violates GDPR by making it hard to opt-out from tracking (which should be opt-in or not at all)... We really need to change the way we monetize internet

According to https://en.wikipedia.org/wiki/Unity_Technologies they're based in San Francisco these days (but were founded in Denmark).

There are protections against that in the EU - you can reverse engineer and otherwise circumvent all the protections for the sake of making things interoperable.

So the good news is that after SV/SF, everything is cheap! Well, mostly....

I would personally suggest Cape Town, South Africa, if you're a bit adventurous. Lovely weather, good people, cheap, excellent food. It's a bit out of the way (10h flying to Europe, more to US), but same timezone as Europe. English is the main language. Moved here 8 years ago and no regrets

Would be really cool if you could just, you know, download the program and run it, not have to go through Viveport at all.

Hi!

Thanks! Feel free to hit me up if you want to chat, some parts of the UI are or can be OS. Email in profile.

Hi

Excellent job! We have been working quite a bit in UI/UX of VR editing. You might be keen to look at what we've done here: https://vrsketch.eu (it's a tool for architects, but shares a lot of problems)

I enjoy how "EU" means "outside of us". I'm in South Africa and it's blocked for the eu

We have been addressing that for the last 3 years. Which CPython modules it does not work with for you?

That comment was about something slightly different though - even though richards is 40x faster, pypy still is not faster on some workloads (like instagram in their measurments), which makes me think that there are other forces at play.

FYI numpy both works these days and it's officially supported

I believe the article is a more or less verbatim transcript of what happened. I wasn't there and I don't really want to speculate, but I would expect LWN to mention everything important that was mentioned and not add their own interpretation either.

The barriers are nearly purely social - the unwillingness to drop C API (or to have a phase out plan) and to declare certain kinds of behaviors as "implementation dependent" make it very hard for any meaningful competition to emerge.

It is harder to make a fast python than to make a fast JS, but it's not that much harder.

It's maybe worth noting that richards is an "easy" benchmark. PyPy has been speeding it up quite significantly by about 40x for a long time.

Those changes are necessary for improved performance, but not nearly good enough. PyPy does that and tons and tons of more stuff and it gets dismissed here as "provided only modest speedup on company workload", which means there is far more involved here than just attribute lookup/function call which have been massively speed up by pypy for quite a few years now.