petschge

20 ms·

181.
by petschge 6y ago
Depends what the authors upload. You can just upload a PDF. But if arXiv detects that the PDF came out of pdflatex it will pretty instantly nag you to give it the TeX sources. It will then render the PDF itself and also offer the sources fo
182.
by petschge 6y ago
I played a bit with the site and liked it. The one killer feature I might pay a few dollars per year for is to create a profile, with a few graphs attached and get a daily or weekly email when a new paper is published that fits well into on
183.
by petschge 6y ago
You get hired permanently if you excel (or promise to excel) in three categories: 1.) Getting grants 2.) Productive science 3.) Teaching. All the skills that the author has gained might help with #2, but he hasn't actually published in
184.
I Regret Quitting Astrophysics(marcelhaas.com)
196 points by petschge 6y ago|140 comments
185.
by petschge 6y ago
And if I go to that website it encourages me to "try it for free" which makes me think I am supposed to pay for it. But ok, let's give it a try. Then I need to enable a bunch of Javascript. Eh, ok. It was recommended on HN, p
186.
by petschge 6y ago
Close. Applying that to double c = dbg(dbg(a) + dbg(b)); gives a = 1 @ debug.cxx:16 b = 2 @ debug.cxx:16 dbg(a) + dbg(b) = 3 @ debug.cxx:16 which has extra dbg()s on the third line. If I could I would probably want
187.
by petschge 6y ago
There is double blind studies that tests efficacy of new vaccines for diseases that already had an approved vaccine. And those studies do find differences in efficacy (just because the old vaccine was 96% effective doesn't mean the new
188.
by petschge 6y ago
Note that FAST was built by China alone, SKA is an international collaboration but without US involvement and NOEMA does have a partnership agreement for data access with the university of Michigan, but is operated and payed for by IRAM, a
189.
by petschge 6y ago
Excess is by definition in comparison to the average of the last few years.
190.
by petschge 6y ago
Yes, there is some changes. Clocks drift, analog receivers age, and so on.
191.
by petschge 6y ago
There was still way more requests for observing time and the total amount of time available in the last cycles. Also there is a lot of value to have long-term datasets that are recorded with the same instruments as it removes a lot of syste
192.
by petschge 6y ago
While maximum resolution is limited by the diameter or the largest baseline in an array, there is one advantage to single dishes: There is no lack of small baselines. Which implies that you can only really get images of large, dim structure
193.
by petschge 6y ago
The cables were (originally) redundant. There are actually replacement cables on the way from Germany right now. But after the one main cable failed earlier this year they noticed that all cables are MUCH weaker than previously thought. So
194.
by petschge 6y ago
Reason 1: because they were afraid of it failing catastrophically just like what happened this morning. One person was on site (director of telescope operations Angel), but imagine what would have happened if a work crew was up in the struc
195.
by petschge 6y ago
The heat capacity of the plasma is actually fairly low. So while it might take a but of wall material, it would cool very quickly, electrons would get captures, and you are left with a bit of gas at rather low pressure.
196.
by petschge 6y ago
I really hope that sympy would improve to a point where I could use it. But as long as Fourier transforms are as badly broken as they are (see e.g. issues [1], [2] and [3]) I can't use it. And the fact that it doesn't fail or cras
197.
by petschge 6y ago
Telescopes are not typically "do nothing. I am not aware of the details of Arecibo, but the big telescope in Greenbank has a paint crew work for 8 hours a day for multiple weeks each summer. And after several years the entire telescope
198.
by petschge 6y ago
You can improve UV coverage a bit but removing redundant duplicate baselines. Which is what pseudo-random patterns do. And these are used e.g. by Lofar. But the problem of lacking very small baselines to fill in very large scales on the sky
199.
by petschge 6y ago
And the US is planning to purchase 2500 of those. Because 2400 are clearly not enough.
200.
by petschge 6y ago
That the US (as of April 2020) does not take any part in.
201.
by petschge 6y ago
Interferometry at optical wavelength is very hard. Doing it remotely is currently impossible.
202.
by petschge 6y ago
Which is why it is good to give absolute risks before and after a change. That way you can easily compute and judge both absolute risk reductions and risk ratios.
203.
by petschge 6y ago
Svalbard might fit the bill. See https://en.wikipedia.org/wiki/Visa_policy_of_Svalbard
204.
by petschge 6y ago
You do. The US does not do "sterile transit". That was a major hazzle for (former) colleagues from South Africa.
205.
by petschge 6y ago
The HPC codes I worked on we would compile with gcc, clang, icc and whatever vendor compiler was installed (Cray, PGI, something even worse). Then we'd benchmark the resulting binaries and make a recommendation based on speed. (Assumin
206.
by petschge 6y ago
Because parallel IO from a lot of different MPI ranks is not supported. And filesystems tend to look unhappy when 100k processes try to open a new file at the same time.
207.
by petschge 6y ago
Having an Xdmf file alongside is nice, but the breaking changes between v2 and v3 are very annoying. And I understand the want to have few external dependencies, but at least HDF5 is straight forward to compile and available as a pre-compil
208.
by petschge 6y ago
Which is why I run valgrind on my code (with a parameter file containing physically valid inputs) to get rid of all undefined behavior. But I gave up on running afl-fuzz, because all it found was crashes following from physically invalid in
209.
by petschge 6y ago
Counter argument: Binary dumps are horrible because usually the documentation that allows you to read the data is missing. Using a self-documenting format such as HDF5 is far superior. It will tell you of the bit are floating point numbers
210.
by petschge 6y ago
Sure that works too. But word of advice from real life: Print the random seed at the beginning of the run so you can find out which seed caused it to crash or do stupid things.

More ›