It’s not that dumb- if a human gets exposed to space the water in their exposed tissues will boil off, leading to evaporative cooling. In a vacuum, evaporative cooling can get you ~arbitrarily cold, as long as you’re giving up enough fluids. I don’t know whether you freeze over or dry out first, but I’m sure someone at NASA has done the math.
The interesting thing about this blog post isn’t the point it makes (which was endlessly made about cafes in all blue-state U.S. cities in the 00s) but that it’s an Iranian writing in English, and thus for an Western audience, about Iranian culture.
What is going on with that? I don’t even dare to speculate. But something more interesting than bougie coffeeshops, certainly.
My primary project the past couple weeks has been applying the AI interpretability technique from Anthropic's famous paper this past summer (you know, the Golden Gate Claude one) to single-cell RNA-seq data. What works on one huge, inscrutible vector space ought to work on another, right? In either case, a fun way to keep in practice, both for comp bio and deep learning.
My side-project, in essence, the bash '&' operator for cases where the first process is already running. It took me months of searching before I could believe that this doesn't already exist, but there you go. I gave in to feature creep, of course, so it's a bit more than that now (I made a ncurses-based dashboard? Why??) but someday soon I'll make it public.
I may be misunderstanding what you're trying to achieve, but you can simulate the bash '&' operator for a running process by pressing ctrl-Z to suspend the process and send it to the background, then running 'bg' to continue the process in the background.
Yeah I was unclear there - I meant the other use of the & operator, 'foo & bar' where bar starts after foo completes. AFAIK, if you ran foo by itself, there's no way to make bar automatically run after foo completes.
The few times I built TUIs with ncurses I wondered: why do I have to program so much by myself?
ncurses is so basic, I didn't have too much fun building UIs with it (more than once). Is there a wrapper or a more modern alternative out there that provides containers and widgets like most GUI frameworks do?
Data scientist and engineer with over a decade of experience centering around biological data analysis and pipeline development. I am equally at home discussing pathway enrichment analysis results from an RNA-seq dataset, as debugging the loss function in a deep learning model, or as architecting a multi-stage processing pipeline on any of the major cloud service providers.
I am Max Alexander, experienced data scientist and bioinformatics engineer, available for short- or long-term contracts. My background is primarily in data analysis and pipeline development for biological data (genomics, transcriptomics, proteomics) and I also hold a Master's in artificial intelligence.
I don’t know whether this is a provocative or tedious thing to say, but the quintessential ‘enlightenmentware’ to have come out of the past several years is ChatGPT. Name anything that brings as much functionality with so simple an interface and so elegant a core!
(It’s simply a pity that we can’t install it locally or tinker with the internals. :) )
> Name anything that brings as much functionality with so simple an interface and so elegant a core!
The most important thing of the last years has been LSP support! I can live perfectly fine without LLM autocompletion (although I did use Tabnine long before ChatGPT came out), but not without a LSP.
The avenue towards making this a viable, routine cure for HIV is by performing a “transplant” of the patient’s own immune stem cells that have been genetically modified to carry the anti-HIV genes. That avoids (if done right) the horrific autoimmune challenges you describe, but like everything else it comes with its own technical challenges that are still being worked out. (I think https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4935568/ is a good review of the whole idea.) To me (though I’m not an HIV researcher by any means!) this sounds like the most promising shot we have at regularly curing the disease.
I’d recommend mynoise.net ; most of the sounds are for-pay but white noise and a few others are free, and it includes an equalizer (and some presets for pink noise and etc.)
It’s not news to anyone that NATO could win a (conventional) war with Russia. The important result is that any sizable chunk of military force in Russia itself stands a good chance of it.
Why risk nuclear armageddon when you can conveniently switch sides at the slightest occasion. It's not like armies marching to shuffle a few chairs in big offices were a matter of deep ideological conflict or worse.
I think you should do some research on the topic before you conclude that "NATO is out of ammunition". I highly recommend Perun https://www.youtube.com/watch?v=deK98IeTjfY .
I'd look to how NATO ran out of PGMs etc when fighting in Libya. I didn't see a big uptick in arms purchases after that. Also, Germany recently crowed about sending 1000 155mm artillery rounds. That's like 3 hours of firing at Ukraine's current rate. The reason Germany has been so slow to provide weaponry to Ukraine is that it is a shell of its former self. The Kriegsmarine has been a joke for decades, the Luftwaffe can barely put 60 aircraft in the air, and the Heer has fewer tanks than Poland, and most of these tanks are in poor condition and shared with the Netherlands.
NATO is not what it was before 1989, and will take decades to come back up to a minimally acceptable level.
There are huge institutional challenges for this to occur. Production facilities have all been consolidated (as they have in the US since the end of the Cold War.) Budgets are also a problem; the majority of NATO members aren't even making the 2% of GDP goal. And there are also some demographic issues; the population of the EU is aging and military service doesn't have as strong an appeal. Will Germany reinstate the draft?
And some things do take decades, even in perfect conditions. Building up a navy takes time due to the long construction timelines for ships. Developing coordination for combined arms operations requires both good officers and non-coms. These take time to develop as well, and need to be continually sustained.
You can set up a schedule to train non-coms in 2-3 months, that doesn't mean they'll be very good. Experience is extremely valuable at this level.
And surely you understand that the US economy and state of industry in 1943 (when the Liberty ships were being launched daily) is quite different than today? Only one country currently has the shipbuilding capacity to even think of something like that. The US (and Europe) have closed down a huge number of shipyards since the end of the Cold War, and it would take decades to create new ones, with trained workforces. All the US shipyards are short workers in almost all categories.
> You can set up a schedule to train non-coms in 2-3 months, that doesn't mean they'll be very good. Experience is extremely valuable at this level.
This one statement means you don’t know what you’re talking about. Non commissioned officers are E4 rank, you can leave bootcamp at E3 rank. 6 months later get E4. NCO only requires experience at the E7 or above level, which don’t fight only lead.
The knowledge is there when he needs it. It comes down from upstairs. They tell him what he needs to know, they tell him what he needs to say, and they feed him all the borscht he can eat. And then there's the extra potato ration on Tuesdays to look forward to.
That's the amusing part. The sad part is that the people behind the troll farms still think they're fooling anybody in the West.
