alexholehouse

Posts like this (and there are many of them) scare me.

I'm in the latter half of a PhD. I love it. I work insane hours entirely out of my own choice, because it is the most rewarding and enjoyable thing I've ever been a part of.

The idea that I've found something that I love, that is challenging, that (I hope) I'm relatively good at, and that has a definite net positive good for us as a species/society, yet I may not be able to pursue this long term because of the immense challenges facing academia as a whole (catalyzed, I would argue, by tragic lack of funding) is really concerning, on both a personal and a societal level.

So the idea that neurodegeneration is essential a manifestation of protein-specific prion disease is not a new idea[1][2][3][4][5] but is one which is steadily gaining traction. That said, from a complexity perspective we're still at the tip of the iceberg.

There's extensive evidence to suggest that the big, fiber-like aggregates that form in these diseases may in fact be neuroprotective, and represent a largely inert thermodynamic end state (basically a big inert crystal which kind of hangs out but doesn't do much damage). More labile, smaller aggregates (basically blobs of protein stuck together) which are formed during the fibril formation process may in fact be more toxic, though again it really depends on who you ask.

How the toxicity is manifest is also totally unclear - is this a loss of function (proteins which aggregate disappear, so they can no longer do their job), gain of function (proteins which aggregate now interact with other things, leading to some new and bad outcomes) or somewhere in between? How does the cellular quality control system interface with all this? Why are certain cell types much more susceptible than others? How does the disease spread in the brain?

There are a lot of very smart people doing some amazing work. If you're interested, off the top of my head I'd take a look at work by Sue Lindquist, Simon Alberti, Marc Diamond, Don Cleveland, Virginia Lee, Eric Ross, Paul Taylor, Rick Morimoto and David Eisenberg (just to get you started!).

The one thing I would say is that I wouldn't treat this as cause for concern for interpersonal spread. If there was a real risk of contamination from (say) surgical instrumentation (as there was with vCJD in the UK) there would be extensive evidence for this (just in terms of a numbers game). Which isn't to say if you put someone's brain in your brain you won't develop the disease, but at that point you probably have bigger issues to worry about...

In my opinion, a much more exciting story relating to ALS was published in Cell a couple of days ago;

Patel, A., Lee, H. O., Jawerth, L., Maharana, S., Jahnel, M., Hein, M. Y., … Alberti, S. (2015). A Liquid-to-Solid Phase Transition of the ALS Protein FUS Accelerated by Disease Mutation. Cell, 162(5), 1066–1077.

[1] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3942086/

[2] http://www.ncbi.nlm.nih.gov/pubmed/24857020

[3] http://jem.rupress.org/content/209/5/889.full

[4] http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3820001/

I'm currently building a 2D/3D simulation engine as part of my PhD - it's uncanny how incredibly simple physical principles generated output that has the "look 'n feel" of organic phenomena.

Related discussion from a few weeks ago;

https://news.ycombinator.com/item?id=9895094

I don't understand;

1) What's the 'top words' which appears when you search for a site? I just get a bunch of profanities (for basically any show, even those which are PG-13). Is this meant to be the top words found in the show's subtitles (it's not) or the most searched for words (in which case why am I'm being shown that). Further searches seem to show some work (e.g. Homeland, or Game of thrones)

2) Expanding the 'top words' gives (apparently) a top 100, except many words appear more than once - in my 'top words' for 'The Simpsons', 'MOM' appears 7 times.

3) What are the 'Top topics'? Again, examining The Simpsons, the top topics are, 'Case/investigation', 'noisey', and 'spooky'.

4) Browser 'back' doesn't work from top topics or top words

Edit: Having read the 'about' I'm feeling far less critical, given this is part of a Big Data course project. Initially, I wondered if the prevalence of profanities in speech (generally) are causing a weird biasing effect (i.e. a single word being said repeatedly) but given there shouldn't be any 'fuck's in The Simpsons/Modern Family/Friends my guess is something may be off on the back-end?

One angle not touched on here is the link between Alzheimer's disease (AD) and sleep. There has always been evidence that AD patients sleep less than the general population, but obviously there's a chicken and egg problem here - are they sleeping less because of neuronal disruption caused by the disease, or is the disease progression catalyzed by a lack of sleep[1-3]? While I think this is still an open question, there's growing research that interaction with sleep related systems in the brain can have a major therapeutic impact on the disease itself - obviously this doesn't actually answer if it's 'sleep' or some process related to sleep, but clearly something is going on here [4].

From the press release associated with [4];

The new research, in mice, demonstrates that eliminating the protein – called orexin – made mice sleep for longer periods of time and strongly slowed the production of brain plaques.

“This indicates we should be looking hard at orexin as a potential target for preventing Alzheimer’s disease,” said senior author David M. Holtzman, MD, head of the Department of Neurology. “Blocking orexin to increase sleep in patients with sleep abnormalities, or perhaps even to improve sleep efficiency in healthy people, may be a way to reduce the risk of Alzheimer’s. This is important to explore further.” - [5]

[1] Xie, L., Kang, H., Xu, Q., Chen, M. J., Liao, Y., Thiyagarajan, M., … Nedergaard, M. (2013). Sleep drives metabolite clearance from the adult brain. Science, 342(6156), 373–377. http://www.sciencemag.org/content/342/6156/373

[2] Ju, Y.-E. S., McLeland, J. S., Toedebusch, C. D., Xiong, C., Fagan, A. M., Duntley, S. P., … Holtzman, D. M. (2013). Sleep quality and preclinical Alzheimer disease. JAMA Neurology, 70(5), 587–593. http://www.ncbi.nlm.nih.gov/pubmed/23479184

[3] Ju, Y.-E. S., Lucey, B. P., & Holtzman, D. M. (2013). Sleep and Alzheimer disease pathology - a bidirectional relationship. Nature Reviews. Neurology, 10(2), 115–119. http://www.nature.com/nrneurol/journal/v10/n2/abs/nrneurol.2...

[4] Roh, J. H., Jiang, H., Finn, M. B., Stewart, F. R., Mahan, T. E., Cirrito, J. R., … Holtzman, D. M. (2014). Potential role of orexin and sleep modulation in the pathogenesis of Alzheimer’s disease. The Journal of Experimental Medicine, 211(13), 2487–2496. http://jem.rupress.org/content/211/13/2487

[5] https://news.wustl.edu/news/Pages/27721.aspx

Technical question - I've seen reports that, "...people who have installed the game to SSD's are apparently getting better performance" [1].

Which implies, to me (both intuitively and having worked on a lot of high performance I/O limited use cases) that part of the issue is read/write speed from the disk? But if that's the case, what kind of software architecture makes sense for a game to have (apparently) real time data processing pulling resources from the disk?

Obviously you need to load maps/games/cutscene etc etc, but this makes it sound like there is constant polling every n-th frame?

[1] - http://www.pcgamer.com/batman-arkham-knights-launch-appears-...

This is such a morbid topic that I've never acted on it, but I suspect that with historical news data you could figure out the growth rate laws which relate time since disaster occurred with number of reported casualties to build a predictive model which - given the inaccessibility and population density of region and the scale of a disaster - would fairly accurately predict the final number of dead.

The logic being that irrespective of final number of dead the 'effective death rate' (i.e. number of new reported casualties per day) is likely to be similar (conjecture).

Anyway - just writing this post makes me feel uncomfortable, hence why I've never done more than hypothesize.

In the biomedical community the standard way to interact with VMD[0] - one of the major molecular viewers and really a workhorse of modern biology - is through Tcl.

I always think this is rather unfortunate, because it means that for many people, their first introduction to scripting ever comes in the form of trying to decipher the Tcl docs and just figure out the simplest things.

[0] - http://www.ks.uiuc.edu/Research/vmd/

Sadly, I think this is perhaps the only (?) reason to stay in academia any more (albeit a very good one).

> explaining that such masses were common among women, and suggested Scott have it checked again in a year.

I know this sounds insane, but menigiomas (which I'd guess is what the neurologist suspected it was) are almost always benign and shockingly common - many post-mortems find these guys just kind of hanging out in/around the brain. The corollary is that any kind of cranial surgery has enormous risks. Point being, if a doctor sees what appears to be a (surprisingly) common event on a chart, they're not going to necessarily recommend further action other than observation UNLESS there's a change to the patient (which, as kitbrennan pointed out, almost certainly accompanied the "get it checked again in a year").

Wait what? They built an 'iOS based operating system'? I don't understand how that would be possible?

Then those people have no business filing software-related trademarks.

Please someone correct me, but assuming you didn't sign anything, I suspect your recourse beyond "AGRGRGRHRGRHGRGHRHGRRHGR" is limited (not that this posts suggests you're looking for vengeance or anything).

That really really sucks.

Original article http://link.springer.com/article/10.1007%2Fs11481-014-9552-2 (paywalled)

As someone who works in dementia research (...), I can't help but notice the total absence of any well defined target for the challenge, which makes it, for a goal-oriented objective process, a poor choice.

"If Dementia wins the vote, the challenge for Longitude Prize 2014 will be to develop intelligent, yet affordable technologies that revolutionise care for people with dementia enabling them to live truly independent lives."

This reflects the fact that there's not an obvious solution to the dementia problem - yes, there are good big picture targets like this, I'm not sure how well this translates into day-to-day operations.

Beyond this, should we be focussing resources/time/money on conditions which, fundamentally, are a byproduct of a high quality of life for a minority of the worlds population, or should we, perhaps, be focussing on ensuring clean drinking water for everyone, or developing antibiotics. I feel like dementia is somewhat incongruous with the other problems presented, but I suspect it might do well because it's so prevalent in the lives of the specific population voting for this.

Or in the shower (or, as I like to call it, "The idea chamber").

This is probably one of my favourite HN posts of 2014. Outstanding.

THIS. One hundred times this. For these results to be in any way meaningful you need to conduct the experiment three times (and, three times with one person isn't actually enough to actually pull statistically meanigful results from, but one step at a time).

1) At t=0 2) At t=x [no transplant] 3) At t=x2 [after transplant]

The thing you actually need to compare is the variation between the difference of 1 and 2 and 1 and 3. The variation between 1 and 2 measures the general variation in your gut biome. So now you have a baseline of what to expect under "normal" variation. So, how does this compare to what you see after the transplant?

Beyond this, how do you control for literally everything in your life around both samples? Maybe you were sick? Maybe you'd been to the gym? Maybe you ate a banana? How do these things affect the microbiome? Without this kind of information its very difficult to draw any kinds of conclusions regarding the impact of a transplant.

As an analogy - lets say I don't know anything about night and day.

I look at my window at some time (say in the morning). It's light. I then write the sentence "LET THERE MY NIGHT" on my wall in marker pen at around 8pm and look at my window. Behold - it's night! So, from this I might conclude writing on my wall brought about the night.

ALL that being said, being curious and engaged in this type of thing is only good. The important thing to bear in mind is that without some thought to experimental design the conclusions drawn cannot really be described as conclusions. This is a great first post, but the determination of natural variation in the gut mirobiome is needed before the impact of the transplant can even be considered.

Original article: http://www.businessinsider.com/ford-exec-gps-2014-1

So I was having a debate with a colleague a few weeks ago: the suggestion was that in 20 years time the a large percentage of surgeries done in hospitals will be done by robots .

Importantly, we didn't specify which surgeries, but through the conversation the implication was not that robots would be removing millions of weird moles and skin marks - we're talking serious, long surgeries.

I'd be really interested to hear the HN community's view on this. I have my own (strong) opinions based on my knowledge of AI and ML as a field, but perhaps other people have their own strong opinions (and perhaps people can change my mind!)

I don't even.... this is amazing. I've been toying with the idea of making something like this, but could never have produced anything as good. I don't know when I was last this excited about a piece of software. When the paperpile option appeared next to PDFs via the Chrome extension - goosebumps! Thank you, this solves a massive problem I have and have had for a while, and despite trying all kinds of managers have never found anything which really combines organization with raw PDF access in a distributed manner.

Evernote + Dropbox means that I am basically device/OS/location agnostic for all my information needs, which given that I switch between 3-4 machines (all with different OSs) regularly is absolutely crucial.

That said, some PAINFULLY obvious things which would be hugely benficial for adoption;

- LaTex support built in

- Markdown (either in every note, or allowing you to add 'markdown' boxes, if just for writing code)

And a stretch goal;

- Better PDF/attachment integration. In a perfect world there would be a Dropbox extension which creates an "Evernote" folder in your Dropbox and any attachments are automatically stored in a notebook-hierarchy file-system within that folder. This would autoupdate to reflect what's going on in Evernote.

It's not Spike's issue, but to be blunt this is a really stupid response. If he really doesn't give a damn he could just say we're looking into it, and if he wants to use this as an amazing opportunity for a bunch of free publicity he could roll with this and take advantage.

This is an interesting question, but could we make this into a real poll?

I'm happy to do this (you need above a certain karma threshold to create polls - I forget what it is) but if the OP would rather not then that's fine too, and I'll leave it.

Consoles are like the archetypical example of loss leader hardware. I assumed this article would be going the other way - that Sony making any profit on the hardware is actually really surprising/impressive.

Holy crap. That's terrifying - where were you when it blew out? How did the staff (pilots and stewards/stewardesses) behave - I've always wondered how well anyone, no matter how much they were trained, would cope in a real [potential] emergency on an aircraft?

That's a really nice idea - and I really hope people will respond well.

On a related note - distrokid looks amazing. If I were still making music I'd sign up for sure, just for the Spotify distribution if nothing else. In fact, I dare say if this had existed 5-10 years ago it would not only have provided a means to push my music out, but more than that, acted as an intensive to actually create and produce.

By lowering the barrier to distribution, you're almost certainly propelling a lot of people to working on musical projects which would otherwise fall by the wayside because there's no obvious end goal.

For some reason I find theories and approaches in compression really interesting. For those unfamiliar I recommend Blelloch's introduction to compression (http://www.cs.cmu.edu/~guyb/realworld/compression.pdf)