cixin
Probably helps if you are also exposed to Kanji on a daily basis in their true context.
Lowering calorie intake is one of the few well known ways that lifespan can be increased. Will have to see if there's any research correlating low calorie diets and telomere length.
http://www.theverge.com/2013/8/28/4669004/google-opens-play-...
I wonder what exactly is going on. Trump administration looking for ways to put pressure on Apple?
If I were you I'd probably do two things:
1. I'd try and save enough money for a couple of years runway in case I wanted to do my own thing/start consulting.
2. I'd look around to see if there are other jobs that might be interesting. My biggest regrets are jobs I've stayed at too long, not the ones I left too quickly. Experience at many companies is great. Now, when you can take or leave a new job, is a good low stress time to look.
Take Psoriasis as an example, doctors are still prescribing coal tar based treatments. I guess they are ~100 years old. They don't work very well, and people don't really know why they should work at all.
Pretty much all Psoriasis treatments are like this, even the more recent ones. They interfere with some pathway related to the immune system... but we don't really understand why they work.
So we don't really understand "how it all fits together" for current treatments. It's still reasonable to use them if they are relatively safe, and work. Same is true of gene therapies.
Mostly we figure out that things work/are safe through experimentation. Not through totally understanding the system.
It's so hard to do anything, even design diagnostic tests, that requires FDA approval that personally I'd avoid any startup that had that on their route to revenue.
It's kind of a shame, and countries where it's easier to develop new treatments and tests are likely to take the lead.
I'm currently using Netflix and iTunes. I also have a prime account, but there's rarely anything I want to watch (the exception being Mr Robot).
I'd love some examples of good plays in this area (that is, those that resulted not just in acquisition but viable products). This area being life science tools/diagnostics.
From the companies I've been it seems to be biased toward crazy plays that had a stack of money thrown at them. Honestly would like to hear about others, know where to read about others.
It seems like there's a pattern of high profile failures of highly funded science-based startups.
From the outside the science seems faulty from the start, the management team seems to have no strong scientific background.
Do investors really fund purely based on their force of personality? Personal connections? Do they see a big market and then ignore the scientific DD?
I'd love to understand this better.
It's both sad, worrying, and slightly comforting that Mirai is run by a bunch of kids. Gives me hope that these issues can eventually be resolved.
But "Thank you all for being a part of this adventure with us!"
Needs to go here:
To be honest, creating multiple new accounts to promote your startup is slightly spammy behavior. Even if they are owned by different employees of that startup.
Just FYI, in the U.K. Non-profits are generally incorporated as companies limited by guarantee. If you are really not seeking profit for shareholders, this would be a better organization type.
But there are two point really:
1. I think it's worth fighting (and they have had some success). 2. May telecoms vendors have volunteered more information than required in the past, I think it's unlikely Apple would.
Moveover, just having a large company challenge these issues changes the political environment.
I think there's value in challenging the US government on these issues. It also helps set the tone in general.
Also, which non-US phone would you suggest? The only real alternative is Android. While there are various vendors, the software is almost completely written by Google. And therefore open to the possibility of a backdoor (or other) requests by the US government.
Android is far for a standard Linux/UNIX system. And it's hard to build one on top of Android because of their use of BIONIC rather than glibc.
So, I feel that Android has taken Linux and crippled it to suit their ends. To me jailbroken iPhones felt much more like standard UNIX phones than Android phones.
My ideal phone doesn't exist. But he iPhone is still the nearest match right now.
Debian is an awesome and somewhat under-rated distribution. Being a maintainer always seems like a thankless and slightly forgotten role. Thanks for having the persistence to keep going so long.
I don't use iCloud for backup. I use my own mail provider. What lock-in are you referring to? I think it would be pretty painless for me to move to Android if I wanted.
I agree the AppStore is more limiting than some alternatives, but overall I think Apple have more interest in user rights than Google.
Android development is not paid for through licensing fees (to any great degree) it's paid for by advertising via Google being the default service/search provider on the platform. This puts you in the position of the advertiser being the customer and the phone user being more like the product.
I personally prefer the Apple model where the phone user is the customer. However I do realize it's not entirely ideal. It just feels like the best option currently available to me.
They present a few cache based attacks. The first is a method to setup a covert communications channel between 2 processes by using cache misses and hits to communicate (basically timing to determine if an address is in cache or not).
Next is attacking crypto. In some AES implementations the T-table sits in a precomputed array in memory. By flushing and probing to see if entries have been added to cache you can find out which parts of the table have been recently accessed. This allows you to extract the key.
I assume you need to probe the whole of memory as you won't know where the table is. I also assume, though not explicitly mentioned that it doesn't matter if you have access to the pages or not, as long as you attempt an access it gets loaded into the cache?
Finally they show that you can monitor keystrokes via the cache, using a method similar to the crypto breaking.
Go feels like a much smaller, similar language though. For me, one of the key features of Go is that it's easy to refactor.
This comes partly from its lack of OO and other features, which are no doubt useful but often create more dependencies between different parts of a codebase.
In my personal programming style, I end up refactoring at lot, Go seems to make this easy.
It's naturally also possible to program like this in other languages. But Go seems to encourage and somewhat enforce it.
Due to a lack of easily abused language features I can also see it being a language well suited to teams where skill levels vary significantly.
Not only do I find it nicer to use (which is somewhat subjective), but I still feel that Apple are standing up for user rights (e.g. FBI/government requests to make phones less secure) more than any other mobile manufacturer.
I've never considered that const might have a performance benefit (and mostly just use it as a compile time check). But will consider it in the future.
Would be interesting to see performance benchmarks for the changes made here, as well as applying const in general.
To recap, GRAIL was created largely by Illumina to help drive sales of their sequencing instruments. Illumina lead (>80% market share) the DNA sequencing market.
However the market is still mostly pretty much research applications, and DNA sequencing has found limited applications in diagnostic settings.
This is their play to start selling into the clinic. Nobody else has really been able to break that market.
1BUSD might finally be enough to make that happen and if it does, that's really exciting. GRAIL are going for screening because that's the largest, most viable market (every adult, every N years).
The problem, is that to be viable it likely has to be cheap. Illumina currently charge about 1000USD per human genome sequenced. Screening costs would probably be similar. Illumina do charge around a 10x markup, so they've got scope to give GRAIL a special deal. But the screening test probably does need to be around 100USD.
If the play works out, Illumina will end up owning clinical diagnostic applications of DNA sequencing. It will kill any other players trying to move into this area.
Really interesting, 1BUSD is a huge bet.
I'm personally pretty happy with that, because for all its faults it does feel like a fun language to program in, for reasons I can't quite put my finger on.
I'm always a little surprised that Java is number one. I know it has a lot of enterprise support, but is there some bias in the survey here?
However, in Bioinformatics, the reality is that many users are still using Perl. It would be get to have them migrate to, modestly more readable languages.
Outside of scripting, alignment and other performance sensitive code is usually written in C (or C++). As with many academic projects, the code quality unfortunately is quite low.