cwyers

36 ms·

391.
by cwyers 7y ago
Manufacturers don't want to put "safe until" on anything because there are a lot of things that can go wrong with an agricultural product and you don't want to provide a firm safety guarantee. The thing is, most labels t
392.
by cwyers 7y ago
> I live in the suburbs and I have to really, really, think. If I want something that isn't deep fried, isn't a burger, or a salad with 'BS' leafs (but is actually a salad with strong content and non-filler material):
393.
by cwyers 7y ago
Getting an ergonomic keyboard near about saved my life. I went from discussing surgery to not even having to be on meds after buying a Microsoft Natural Keyboard.
394.
by cwyers 7y ago
There are alternatives -- Julia and R both have bindings for Keras, which gets you Tensorflow, CNTK and Theano. (I think there are bindings for Tensorflow directly from R and Julia as well.) Once you have a trained model, it doesn't re
395.
by cwyers 7y ago
I once spent a bit of time playing with DataFramesMeta.jl, which is what I was able to find that seemed like Julia's answer to dplyr, and I thought it was awful precisely because everything was macros. With dplyr, any function that tak
396.
by cwyers 7y ago
> But that’s just the first ten years. After that, nobody can think of a single feature that they really need. Is there anything you need that Excel 2000 or Windows 2000 doesn’t already do? With all due respect to my friends on the Offic
397.
by cwyers 7y ago
Building mature software can take a long time, I agree. It depends a lot on the complexity (an operating system is obviously more complex than a solitaire game). The part I disagree with is the idea that you get to done. Photoshop is a grea
398.
by cwyers 7y ago
I don't actually think that aged very well. Joel's company has been renamed after a product that's two years old, and his mature product has been fading in the market as newer alternatives like Jira and Asana and (ironically)
399.
by cwyers 7y ago
Azure Hybrid Benefit is for licenses with Software Assurance. The change that the blog post keeps railing on about is for licenses without Software Assurance.
400.
by cwyers 7y ago
Yeah, Hipchat is EOL and Teams has a market. Now, I _hate_ Teams, but it is nonobvious that Hipchat has a better notion of what features are necessary to be successful.
401.
by cwyers 7y ago
Yeah, "Organization" includes Basecampe, Airtable, 1Password, Github and Google Drive. At that point, what isn't organization? EDIT: That said, I see what they're getting at with putting Preface into project management.
402.
by cwyers 7y ago
> And that is precisely why I don’t see how ML algorithms are “more powerful” than statistics in any way, as per your claim. Well I'm glad you don't see that, but I'm a bit confused why you think I said it because I didn&#
403.
by cwyers 7y ago
I don't think random forests are "magical" at all. I think you can walk a non-technical audience from decision trees to bagged and boosted forests on a whiteboard in under an hour. (I don't think you can do this nearly a
404.
by cwyers 7y ago
I think what you're saying is orthogonal to what I'm saying. Yes, ZIP codes can be a proxy for race in models dealing with credit scores (and recidivism, which is also a really bad place to put racial bias), as an example. But if
405.
by cwyers 7y ago
Everybody owes it to themselves to read Breiman's Two Cultures[1]: > There are two cultures in the use of statistical modeling to reach conclusions from data. One assumes that the data are generated by a given stochastic data model.
406.
by cwyers 7y ago
The big change since then is that Debian switched back to ffmpeg: https://lwn.net/Articles/650816/
407.
by cwyers 7y ago
I think the word "culture" is overloaded, such that it's really hard to talk about and you can make it mean whatever you want, but there's a use for it that I don't know a better way to describe in a word. Yes, ther
408.
by cwyers 7y ago
So, I think there's two antipatterns and a pattern here. Antipattern one is the one where you just get on the treadmill and churn, baby, churn. Constantly replacing things, spending way too much resources on tech and not enough on meet
409.
by cwyers 7y ago
The article mentions this: > I know you love programming because you like technology, so this may go against your very nature, but no one says you’ve got to jump every time some snot-nosed kid invents a new way to run byte-code. You have
410.
by cwyers 7y ago
Yes. And the paragraph that most stood out to me was this: > If the task you are working on can be expressed as a pure function that simply processes input parameters into a return structure, it is easy to switch it out for different imp
411.
by cwyers 7y ago
I also am trying to figure out how this stacks up against Apache Drill.
412.
by cwyers 7y ago
Interesting that they opted for the single-most popular query language on the planet, versus somebody's hobby project? Why is that interesting?
413.
by cwyers 7y ago
> Bayes' rule is a great tool in many circumstances! But it has a great weakness: it requires a prior. That doesn't make it useless; few tools are useful in all circumstances. But requiring "everyone to use Bayes' rul
414.
by cwyers 7y ago
Where are you getting this from? The actual test scores in the graph in the article don't correspond with the percentiles; I don't think the scales on the x and y axis are the same, which is what you seem to be contending.
415.
by cwyers 7y ago
Nobody said it had to be scale-end effects alone. But yes, when I look at that Dunning-Kruger graph, I don't see what the authors of TFA say or what the common misinterpretation is (I agree with the authors about how it's commonly
416.
by cwyers 7y ago
Why not just use Linode or Digital Ocean at this point?
417.
by cwyers 7y ago
Yeah, watching Hacker News try to violate sanctions from first principles is amusing.
418.
by cwyers 7y ago
Advertising is not immune from market corrections. The whole "pivot to video" trend was a market corrections to ad rates, _and_ has died off in a response to another market correction. https://en.wikipedia.org/wiki
419.
by cwyers 7y ago
Which lasts until the advertisers figure out that the views are going to toddlers watching creepy videos featuring unlicensed Disney IP and bad Peppa Pig creepypastas, and the party ends.
420.
by cwyers 7y ago
I'm like, isn't this old news? But no, that was Purdue: https://www.motherjones.com/politics/2019/01/report-mckinsey...

More ›