ms013

I've moved to doing ocaml development on my windows 10 machine under the windows subsystem for Linux. The Linux ocaml tool chain works pretty seamlessly there. Works very smoothly relative to past experiences with native ocaml compilers on windows.

You didn't read the article apparently. Fortran (which officially dropped all caps in the standard in 1990) has had new versions every couple years, up to 2013. Each revision of the language brings modern features into it, and old features go through a process of deprecation and then obsolescence. Most people seem to be unaware that it is an evolving language that has changed significantly since the standard most people are familiar with from 1977. I am a bit biased though: I work on a Fortran compiler, so I've had reason to stay on top of the standards as they come out.

They epoxied the USB ports on my secure machine when I worked at one of the national labs about 12 years ago. One day I came in and the admin had gone through everyone's office the evening before with a bunch of JB weld to fill the USB ports.

So, whether or not you believe that it is a useful solution, it certainly did happen. As for alternatives to gluing, I assume that was time related: filling in a port is certainly faster than opening thousands of machines to give them a physical port-ectomy.

Finite element methods are often used to validate that a structure meets its expected requirements, and are part of the CAD workflow. The Sleipner-A offshore platform failed and sank due to bad finite element analysis.

See: https://www.ima.umn.edu/%7Earnold/disasters/sleipner.html

For those who have Mathematica and want to experiment with this, here's a quick function to generate the data:

    f[n_, base_] :=
     Module[
      {m, d, dpairs},
      d = Table[Last[IntegerDigits[Prime[i], base]], {i, 1, n}];
      dpairs = Table[{d[[i]], d[[i + 1]]}, {i, 1, Length[d] - 1}];
      Map[#[[1]] -> #[[2]] &, Tally[dpairs]]
     ]

For the first n primes in a given base, it returns the mapping {i,j}->count for the all pairings of digit i followed by digit j. E.g., for the first million base 5 primes

    {2, 3} -> 68596
    {3, 0} -> 1, 
    {0, 2} -> 1, 
    {2, 1} -> 64230 
    {1, 3} -> 77475 
    {3, 2} -> 72827 
    {2, 4} -> 77586 
    {4, 3} -> 64371 
    {3, 4} -> 79358 
    {4, 1} -> 84596 
    {1, 2} -> 79453 
    {4, 2} -> 58130 
    {4, 4} -> 42843 
    {1, 1} -> 42853 
    {3, 3} -> 39668 
    {2, 2} -> 39603 
    {1, 4} -> 50153 
    {3, 1} -> 58255

> > Only losing market share and power has made Microsoft produce software with reasonable quality, flexibility, and interoperability.

> And that, in my opinion, is the way it's supposed to work. Vendor's products gettin' ya down? Not being responsive to your needs and lacking in quality? Move elsewhere! Money talks.

Exactly. That's what got Apple where they are today as well. Pre-Rhapsody, MacOS was largely a joke in the developer community, but that changed drastically when they ditched the archaic MacOS for the Nextstep derived OSX. When that happened, suddenly there was a reasonable alternative to Windows or the various Unices out there. It seems entirely natural that, like the shift that pushed people towards Apple, we would see similar shifts in other OS's that would entice people to think more broadly than seeing Apple as the dominant desirable dev platform. That seems like a healthy thing with respect to avoiding a stagnant monoculture.

The other fun one is people who refer to the input files for the simulation packages as "input decks" (aka, punched card decks). I occasionally get a funny look when talking to someone from outside the computational science community use that term. And I'm in my 30s - punched cards were history by the time I started using computers myself, so this kind of terminology isn't really generational - it seems to come from specific communities.

I personally have mixed feelings about Mathematica. Stephen Wolfram has issues, but I don't buy software based on the attitude of the creator. Otherwise, many open source projects that I use with some combination of very cranky, irritable, rude, egotistical, and/or arrogant people involved would be ruled out as well...

Closed source that costs $$$ is definitely an issue - for a while after school it was hard to continue using Mathematica until my employer acquired a site license and I got a home license. In the meantime I tried all of the usual suspects - from open source things (Maxima, Sage, Octave, Julia, and so on...) to other commercial products (e.g., Maple and Matlab). I ended up returning to Mathematica for two reasons: 1) it has a massive standard library, and once you learn the idioms of the library and the language, I found it pretty usable; 2) it has beautiful graphics capabilities.

The open source tools are hit-or-miss on the standard library front - some of the libraries are great, some are missing major features, or their implementations are poor. (Yes, it's OSS and one can either contribute code or feedback, but often I have a problem to solve NOW - I don't have time to wait for the OSS to catch up.)

As for the graphics, I'm not sure what it is about Mathematica visualizations that I like so much, but they look GOOD relative to the others I've tried.

At the end of the day though, Mathematica is just one of the useful tools in my toolchain - the others have their plusses and minuses as well. But one thing I've adapted to over the years is that A) sometimes good software costs money, and B) sometimes jerks are involved in creating good software. Neither case is one that I feel is a valid justification for rejecting a software package outright.

Edit: It's worth noting the time period over which I've been mucking around with Mathematica and the other alternatives I mentioned above. I started seriously using numerical and scientific computing software (including Mathematica) in 1994, so that's a 21 year period of wandering around the space trying out things here and there.

I'm not one of the org-mode users. I've been a big fan of Onenote for a while - works great on my phone, mac, and windows computer. I tried org-mode when one of my co-workers kept talking about it, but it didn't really work for my workflow, even though I use emacs for most of my code-related tasks. Just use whatever works best for you - if you want to use Onenote, it doesn't really matter if org-mode or some other tool happens to be more popular in some community.

Air entering the engine at high speed is heated significantly due to compression.

Looking for people in the Portland Oregon area. I have an idea that has been bubbling for a while to change how scientists use and track computation as part of their work. I have various components prototyped, but lack front end and DB experience. I've got about 15 years experience in this area, and have been getting the startup itch lately to make some of the ideas materialize.

Hawaii. Rent a condo/house on VRBO, that way you are not stuck at a resort with tourists. Typically can find these near resorts, so you are close enough to enjoy them when you want to close the laptop. I've had some good weeks of hacking between snorkeling and sunning on Maui, renting a condo on a golf course near the resorts north of Lahaina. Kauai and the Big Island have similarly nice places with reasonable housing and food, with the ocean nearby. Avoid the resorty areas of Oahu.

As for outdoor wifi, often just making sure I'm in a spot with LTE connectivity is sufficient to live with tethering to a phone or my ipad to do basic net things like access repos, web pages, and occasionally download packages.