stochtastic
Joined 9 years ago
Karma
227
Submissions
34
Member since
Mar 7
Recent submissions
Comment · 8 months ago
Why? It's an excellent recruiting tool. I used to read it as a kid (along with every other paper or digital encyclopedia I could get my hands on), and it certainly made me interested in the CIA.
Comment · 1 year ago
This is very cool, thanks for sharing. There are many industries and academic fields where Matlab has gone out of fashion (I work in one), and my experience has been that there's a lot of over the top negativity about Mathworks products, centering mostly on the license and business model. But I think there's a lack of awareness of just how superior Matlab and Simulink are to all alternatives in some domains. All that to say: don't let the Matlab hate get to you, you're using the right tool for the job.
Comment · 2 years ago
This is a fun dataset. The paper leaves a slight misimpression about channel statistics: IIUC, they do not correct for sampling propensity to reweight when looking at subscriber counts (it should be weighted ~1/# of videos per channel since the probability of a given channel appearing is proportional to the number of public videos that channel has, as long as the sample is a small fraction of the population).
Comment · 3 years ago
It also includes the Apple One bundle. I wonder if the bundles complicate that breakdown — I certainly never use Apple Music or Arcade, and almost never TV+.
Anecdotally, I cannot figure out how to cancel One without losing my iCloud photo library. As soon as I do figure that out, I'll be iCloud only.
Comment · 3 years ago
I don't think they were pretending to look, they were just (rightly) remaining open to it being either rescue or recovery. From the article:
> “While not definitive, this information was immediately shared with the Incident Commander to assist with the ongoing search and rescue mission.”
So:
* They knew about it and quickly aided in the search.
* They did not broadcast the details of the collection to the public.
That seems like the right call to me.
Comment · 3 years ago
> and so must be hiding that they can do so
I don't believe I said that. You can draw your own conclusion from the fact that it is within their capabilities to detect, localize, and to some extent classify a wide range of sources in this region of the ocean.
What you won't find is a lot of information about those capabilities in the public domain. Just consider that what _is_ known tells us that we had these capabilities in the 1950s, and that they were continuously improved upon throughout the cold war. This is not Area 51 conspiracy speculation; it is bread-and-butter NRL stuff that is more than half a century old at this point and is classified for good reasons.
Comment · 3 years ago
Hydrophones don't need to be nearby or even at a similar depth. The SOFAR channel acts as a waveguide and will duct sources from other depths as long as the bottom is below the critical depth. As others have said, this part of the North Atlantic is one of the most heavily monitored parts of the ocean as well. No sci-fi physics necessary — this has been done continuously since the 1950s.
Comment · 3 years ago
> We've gone from: - Uncertainty exceeding signal in satellite data is "simply not the case" ...
Not sure I can help someone who willfully misreads my comments! Give chapter 9 of the AR6 a read and then we can talk.
Comment · 3 years ago
The thread is at max depth, so I'll reply here:
As you quoted from the paper:
> The six main groups that provide satellite-altimetry-based GMSL estimates ...
1) You're restricting your attention to groups producing satellite-derived estimates.
2) Several of those groups, e.g. CSIRO, are using data assimilation to combine satellite estimates with gauges [1].
3) There are multiple satellites, often overlapping in time, often of totally different design/orbit/etc... [2]
4) Up to the most recent IPCC report (AR6), assessments were based on tide gauges alone [3]. For many of the reasons that you bring up. Hence "Nobody relies on estimates from a single satellite alone." Reconstruction of historical sea level is a huge scientific discipline and you would benefit from doing some more reading before beating a dead horse. The estimated uncertainty is large for a reason, but it does not encapsulate any scenarios where the observed change has been insignificant. If you think otherwise, I'd be happy to take your money for 2100 options on coastal real estate in about 90% of the world's coastline.
[1] — https://research.csiro.au/slrwavescoast/sea-level/measuremen... and https://link.springer.com/article/10.1007/s10712-011-9119-1
[2] — https://research.csiro.au/slrwavescoast/sea-level/measuremen...
[3] — https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6... p.1287
Comment · 3 years ago
Figure 1 from that paper didn't convince you that it was a minor issue? Their correction is about 7% of the 1993-2018 signal, and it resolves the discrepancy with respect to other estimates, including non-satellite ones. Nobody relies on estimates from a single satellite alone.
https://essd.copernicus.org/articles/11/1189/2019/essd-11-11...
Comment · 3 years ago
> Also that assumes the scientists are completely accurate and don't make any mistakes. As recently as 5 years ago they discovered that the measured rise for almost all of the 90s was wrong and revised it by 3mm/year +- 1.7mm/year - the error was the same amount as the imputed level of rise!
Citation for this? There are always improvements to our understanding of past data, but you seem to be implying that the uncertainty exceeds the signal. That's simply not the case [1].
[1] — https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6... pp.1291-1292
Comment · 3 years ago
Archimedes would like a word. There are some small effects due mostly to salinity differences, but sea ice isn't a major contributor because it is already floating and thus displacing an equivalent volume of water:
Comment · 3 years ago
This article only briefly touches on what I think is the most interesting aspect of sea level rise: gravitational adjustment. Even though I never worked seriously on sea level research, it was by far my favorite course in graduate school. For anyone who would like to learn more, I highly recommend listening to Jerry Mitrovica expound on the subject [1]. I guarantee that you will learn a few things that will stick with you.
In my experience, the first order adjustment effects are not widely known even among physical oceanographers. I remember describing it to a preeminent expert on El Nino, and him being incredulous that local sea level fall near the source of meltwater could be the first-order effect beyond accounting for rebound. "The first-order response is a Kelvin wave — you've surely been misinformed."
[1] - One example lecture: https://www.youtube.com/watch?v=s_rGksRoG9A
[2] - Another, shorter summary talk https://www.youtube.com/watch?v=yj7JIxU9Af4
Comment · 4 years ago
I've been very impressed with sqlglot, and am looking forward to trying this feature. The only issue I've had with sqlglot is transpiling for use with a specific spark version: in my experience Spark is not great about surfacing obvious 'not registered' errors when a function isn't supported (especially in >=2.4). I ran into this with width_bucket, which is only in the most recent release. I am curious whether there's a straightforward way to write with a specific release and catch the error in transpilation rather than execution.
Side note: Iaroslav (post author) and Toby (sqlglot creator) are both amazing, and I'm so glad that they're working on open source projects like this.