fijal

20 ms·

121.
by fijal 11y ago
I wonder if it's just me or does the complaint concerns mostly oc cupid, which is very data obsessed. For example Tinder, a far more successful online dating tool seems to be much much simpler in purpose and not suffering from those pr
122.
by fijal 11y ago
pypy3 is quite a bit slower than pypy for 2.7
123.
by fijal 11y ago
Note that 50 mile in 20min is a (relatively modest for european standards) 240 km/h. The fastest links in europe can sustain 270-300km/h, using traditional, overground trains. Since you would probably need to dig, you don't c
124.
by fijal 11y ago
Hi The python code that's generated is ok, but the problem is that everything is very dynamic. If you want this to work fast on PyPy, it should really generate a bunch of getters and setters that use attribute access and not go through
125.
by fijal 11y ago
there is no wrapper ;-) at least no wrapper generator, we just hand-wrapped some stuff.
126.
by fijal 11y ago
Hi. We had tons of success using protobuf-c and wrapping the resulting wrapper using cffi. The pure-python protobuf wrapper generates really terrible code - don't use it for anything performance-sensitive.
127.
by fijal 11y ago
I think a bad sentiment here is mostly generated not about what Mark says on his post, but what he does otherwise. We've seen a fair share of controversy regarding facebook, privacy practices etc, we've seen a fair share of contro
128.
by fijal 11y ago
They completely brought it upon themselves. You can't just release a phone and then essentially abandon your users - the phone is buggy, email client and browser are borderline unusable and they break features in between releases. It a
129.
by fijal 11y ago
I think there is a distinction here between "groupon customer" and "business customer" highlighted in that comment that goes beyond the term "bargain hunter". You can still have bargain hunters that would only
130.
by fijal 11y ago
Hi Andrew, thanks for taking your time. Since you're trying to capture it in one place, there is a comment hidden down with ctiticism that you did not capture in your list, here it goes: "unclebucknasty 3 hours ago That and the fa
131.
by fijal 11y ago
I like how the article start with making fun of ancient body measures, but then uses feet to compare.
132.
by fijal 11y ago
Wow. 125bln USD is half of greek GDP and just under a third of its debt.
133.
by fijal 11y ago
If you can change the extension, then going for cffi seems like a no-brainer. It works on cpython, it's easier to use, safer etc. Usually the problem is if the application/library is too big/too cumbersome. Then I think i
134.
by fijal 11y ago
No, mostly business. Those things are typically not criminal so as far as open source goes for the sake of open source, it's easy to ignore
135.
by fijal 11y ago
I would sign up my project for it, if it can generate more revenue than is worth the effort
136.
by fijal 11y ago
Uh, the lack of email was an omission - fixed. I worked for the past few years on making PyPy & satellite projects more business friendly. We've established a company (baroquesoftware.com), I went to conferences, advertised it etc.
137.
by fijal 11y ago
Hi toyg. You clearly haven't tried raising money for an Open Source project. It's really really difficult and the amounts of money you get range a lot. Usually you also want to underpromise and overdeliver, so it's about rais
138.
by fijal 11y ago
Actually pypy tries to do the same, bitbucket does not support tagging very well though (but looking for "easy" or something in the issue tracker helps). I am personally sorry if the poster had an experience like that, IRC is diff
139.
by fijal 11y ago
https://morepypy.blogspot.pt/2015/10/pypy-memory-and-warmup-... https://morepypy.blogspot.pt/2015/09/pypy-warmup-improvement...
140.
by fijal 11y ago
yop, as I said, ujson is probably your best bet if all you do is parse json.
141.
by fijal 11y ago
we did kickstart the python 3 support on our website. Donate!
142.
by fijal 11y ago
Generally speaking "no", but there are a few cases where it might, e.g. if you're using arrays/numpy arrays in list comprehensions. We could not get it to run on normal lists for a bit obscure reasons.
143.
by fijal 11y ago
Try builtin json instead of simplejson for pypy (it's API-equivalent)
144.
by fijal 11y ago
Working on it (even wrote 2 blog posts recently about it). Mercurial is an interesting benchmark/use case for us.
145.
by fijal 11y ago
We did put some effort into making built-in json faster. If you're using an optimized version like uJson, it'll be faster under cpython so far, but it all depends whether there is more processing or more parsing involved. We'
146.
by fijal 11y ago
we're trying to address both numpy and C extensions, stay tuned ;-)
147.
by fijal 11y ago
yes, but BLAS only helps you with some operations and not others. In fact most of "basic" operations are still done "by hand" in some templated C in numpy. Additionally, doing vectorized operations lazily (which is not p
148.
by fijal 11y ago
I have a question to all the naysayers and praisers of the future autonomous car: have any of you ever been on a high speed rail link in europe? Assuming everything else equal, they're just a far superior way to travel, where you can w
149.
by fijal 11y ago
I'm not really going to argue with you, if you can make numba perform - great! We're aiming at providing middle ground. Situations where the code is heavy on numerics, but complex enough that numba either can't handle it at a
150.
by fijal 11y ago
It's maybe worth noting that we (pypy team) had a huge success testing C code just with cffi, writing tests in python. It gives you the ability to write C, but with all the flexibility of python for testing and some debugging (sometime

More ›