Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
pwang
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
18 ms
·
31.
▲
by
pwang
8y ago
Beautifully put. We bootstrapped a global inter-network pretty far on the basis of the "Personal Computer" paradigm, but are running into the limits of that at every level. Servers and server farms comprised of PCs are horribly c
32.
▲
by
pwang
8y ago
Exactly right. There is an irreversible entropy to this: once a certain number of implementations exist, it becomes almost impossible to create a "new standard" that fixes them all, because it has to be backwards compatible with
33.
▲
by
pwang
8y ago
* The package formats (source or wheel) do not sufficiently capture assumptions that devs make at release time, in order to produce a usable installable artifact at install time. * Worse, when different package builders release binaries, th
34.
▲
by
pwang
8y ago
> You can get all the packages you need for web development, AI or statistical/mathematical stuff in python from say debian packages Um... not quite that simple. If you are just developing for yourself, and you have full control ov
35.
▲
by
pwang
8y ago
The heart of the problem is that different people are doing different things with Python, and nobody wants to pay cognitive load for what they don't want to use. The packagers and package tool makers of the world inherit ALL of the tec
36.
▲
by
pwang
8y ago
What if the right answer to the packaging nightmare that plagues languages with native extensions, is to actually have multiple isolated copies of the interpreter, each with well-known build chain metadata and sane, isolated locations for s
37.
▲
by
pwang
8y ago
https://bokeh.pydata.org/en/latest/docs/gallery/hexbin.html Write Python, make interactive web plots.
38.
▲
by
pwang
8y ago
You mean like https://demo.bokehplots.com/apps/gapminder ?
39.
▲
by
pwang
8y ago
> If you want to do dynamic visualisation with something like Python or R in a browser you need need a server, and that incurs significant latency. Not true. Bokeh ( http://bokeh.pydata.org ) lets you build interactive web gra
40.
▲
by
pwang
9y ago
Agreed! Empire Deluxe is my favorite! Sadly, I've never been satisfied with the Killer Bees "modernization" ports - the UI/UX is odd and deficient. I just want a good Empire Deluxe for tablet. Polytopia is the closest
41.
▲
by
pwang
9y ago
Holy shit. ScriptedForms is amazing. Is there any possibility you would consider selling a dual-licensed version that is not AGPL3?
42.
▲
by
pwang
9y ago
See Myth #6 here: http://jakevdp.github.io/blog/2016/08/25/conda-myths-and-mis...
43.
▲
by
pwang
9y ago
This is also a useful read: http://jakevdp.github.io/blog/2016/08/25/conda-myths-and-mis...
44.
▲
by
pwang
9y ago
Anaconda does not leverage virtualenv under the hood. Anaconda does not use wheels as part of that. However, you can use pip and virtualenvs within Anaconda, but you'll probably get some inconsistent results. The reason that Anaconda
45.
▲
by
pwang
9y ago
Pipenv is only as good as the wheels that get installed. And the wheels are only as good as the assumption that your runtime/deployment environment matches the buildtime environment of the person who made those wheels. Oftentimes wi
46.
▲
by
pwang
9y ago
What do you mean by "same OS requirement"? Can't you take your environment.yml from a Mac and recreate the environment on Linux?
47.
▲
by
pwang
9y ago
Any cloud computing environment focused on data science or machine learning will have Anaconda installed by default (and for good reason). The sheer amount of complex dependencies and possible errors between build-time configuration and run
48.
▲
by
pwang
9y ago
Many people who do data science on OS X have to deploy or scale their things on Linux servers. Or serve Python scripts to collaborators on Windows. Even if you are a working on an island and you fully control your own dev and production en
49.
▲
by
pwang
9y ago
Have you actually done this? If so, on what platform(s)?
50.
▲
by
pwang
9y ago
Yep, that's why I like it. :-)
51.
▲
by
pwang
9y ago
I think JS is critical infrastructure to this because the only way this stuff takes off is if users flock to it. And so the UX can't regress too much from what they're used to, and the modern web is what they're used to. In
52.
▲
by
pwang
9y ago
From what I've read and seen in the docs, faq, presentations, etc., it appears to me that IPFS really pulls in a lot of stuff about the network layer, because it's trying to be, well, an Interplanetary (distributed) File System. D
53.
▲
by
pwang
9y ago
From what I can tell (and I am not an expert): Zeronet and IPFS are really fixated with solving the P2P secure content distribution problem. They convolve the solution of that transport-level problem with what I believe to be the central p
54.
▲
by
pwang
9y ago
P2P and anonymity are orthogonal concerns. Maybe I want to peer-wise distribute content to well-known friends of mine, to whom I am well-known. Alternatively, perhaps I want to contribute an anonymous voice to a globally centralized conve
55.
▲
by
pwang
9y ago
- DAT is not about decentralization; it's about content integrity - Beaker is an app framework built on Electron/Chrome because without gorgeous apps you don't get users. But all of the data is stored within DAT, and there&#
56.
▲
by
pwang
9y ago
"The only reason to pick python is momentum, libraries and mindshare." I'm confused. Are you using this as an argument for Julia? I don't post "LOL" very often on HN, but congrats... LOL! Unless you code in a
57.
▲
by
pwang
9y ago
The whole point of programming language conferences is to focus on bleeding edge "what's new/potentially coming down the pike", instead of "what's being adopted massively". Of course the type of people who
58.
▲
by
pwang
9y ago
I believe much similar magic is possible in Python, but you get to implement it on top of the relatively simple interactions between instance dict, class dict, and metaclass. They would not be bullet-proof, and they are not built into the
59.
▲
by
pwang
9y ago
When you use numpy, you're not one semicolon away from a segfault.
60.
▲
by
pwang
9y ago
When Python is being used for these data analytical tasks, random attribute access is not the common case. Instead, SIMD and vectorized mathematical transformations is the order of the day, and those can be specified very concisely and p
More ›