We just had the Chinese government telling us (essentially) to shut down free speech and eliminate the multi-party system.
Now it's the corporate world's turn to tell us what to do.
748 karma · joined October 4, 2015
Now it's the corporate world's turn to tell us what to do.
I had seen a bunch of other Gaudi buildings, and wasn't all that impressed. The outside of La Sagadra Famailia just seemed tedious.
I went inside and ... words fail me to describe the experience.
I was in primary school (age 11) and we had a Microbee at home with 64KB of RAM and a copy of Turbo Pascal. My sister is/was dyslexic so with a bit of help from my older brother (who would have been in first year of computer science at the time) I wrote a spellchecker for her. It think it might have been a Christmas present, so I had probably been programming for less than a year at that stage, and only in the afternoons after I got back from school.
It read the whole document into memory (not the English dictionary, which was too big) in a tree structure. Pointing out that I could use a binary tree and teaching me how to use pointers was my brother's contribution. Then it read through an on-disk copy of the English dictionary. It wasn't a complete dictionary, so I didn't do any compression on the storage, and then asked the user (my sister) to review each word that wasn't in the dictionary, sending any issues out to the printer with a line number and the word in bold.
So if a kid with very little experience could do it, professionals would have had little trouble back then.
Writing a spell-checker on the computer we had before that though (the ZX-spectrum with 16KB of RAM and the only bulk storage being a cassette tape)... that would have been hard. We've definitely progressed since then.
If you are a large-ish business, you will talk to your insurance broker to insure it. This will probably turn into a multi-thousand page document describing all the assets that you want protected, and what events you want them protected from.
The broker then does "marketing" which means nothing like what marketing means in any other context. It means that they are going out to the underwriters and asking them what parts of that whole package they are willing to underwrite, and at what costs.
The broker then cherrypicks whatever is best for their client by piecing together underwritings for different parts. ("Allianz is underwriting the buildings for the first $800m, Lloyds will do everything beyond that up to $3B, SwissRe had the best weather insurance...") The client wants to know that the broker genuinely did approach as many underwriters as they claim and validate that the underwriters did reply with the quotes that the broker claims were replied. (This is how the client makes sure that the broker and the underwriter aren't colluding.)
The end result is that across the industry there is a many-to-many broadcast and scatter and gather that needs to happen and be notarised. (The broker sending stuff out, the underwriters responding.)
At the moment, there is usually a third party hub through which these documents flow who essentially notarises the communications, but also has no incentive to allow collusion between parties. So everyone trusts that third party.
The only problem is, there is no incentive for the third party to be efficient, and so in every country / state / market where this kind of system operates, the third party inevitably takes a large cut of fees and provides quite remarkably terrible service. (Case in point: "of course it's secure, it's on port 443". "Umm, you are actually just making an HTTP connection on port 443, there's no SSL happening on this connection, as can be seen in this packet trace.")
The brokers hate how it has played out, the underwriters hate it even more. With a bit of handwaving, this is something that can be solved by a blockchain. Although the unencrypted documents themselves probably can't be on the blockchain, the checksums can be, and maybe there's a scheme where the encrypted versions of documents can be there too.
It's one of those situations that blockchains are good for: distributed system, no trust required, the users of the blockchain can afford the appropriate technical skills to keep the system running, can't be replaced by an independent 3rd party, etc. etc.
But it's not likely to happen.
The main problem is that there's no monetary incentive for any individual party to build the system. It's a tragedy of the commons problem: once it is in place, it will get used, but no-one wants to pay for it because if someone else can pay for it first then that's a better arrangement.
So yeah, five years of discussions and learning about the market, and everyone wants it to exist but no-one wants to pay.
But VMS clusters were much smaller, so the complexity of the distributed key stores were much simpler.
- With threads, one thread can take the others down. With OS processes, it's hard to keep track of which ones are alive and available to send to. With (most) async programming models (e.g. Javascript, Tcl) you only end up utilising one CPU. With Ada-style async programming models you have to think hard about deadlocks and mutexes. The BEAM model solves all these problems.
Who gets the slots available in a course is dictated by the students' results in the final exams in high school. (This is approximately true: in NSW several exams before the end of high school also count. Some universities will offer places based on the "other" high school exams if they can predict what your final exam result will be with sufficient accuracy.)
The student with the highest marks gets the place they want, and then the next student down gets their first preference, unless it is full, in which case they get their second, third, fourth, preference. And so on down through all the student candidature.
Theoretically this whole system should let a sufficiently capable average income student study humanities (or whatever they want).
In practice there are problems -- some schools have teachers who are very good at getting good marks for their students in the final exam; actually being intelligent and creative doesn't really help as much as it should -- but it's not an overly terrible system.
In the USA, tax deductions to churches is giving the same treatment as a charity.
I was mostly profitable by about week 5 -- so I only skipped one pay cheque, but I didn't have children then, so it was easier to live cheaply.
But that was a service business: service businesses are easy to get started, but they are hard to scale. If you are thinking of making a product business, you are in for a much longer harder road.
It's unlikely that your users are going to notice the accuracy difference between the linear model and the GPU-intensive one unless you are doing computer vision. If you have small datasets, you might even find the linear model works better.
So it won't affect revenue, but it will cut costs to almost nothing.
Supporting evidence: I just completed this kind of migration for a bay area client (even though I live in Australia). Training (for all customers simultaneously) runs on a single t3.small now, replacing a very large and complicated set up that was there previously.
Machine learning: what happens when we replace Euclidean metrics with p-adic ones? Distance is fundamental to so many algorithms (least squares regression; nearest neighbours; anything involving gradient descent). How do those algorithms behave over completely foreign metric spaces?
One day on a whim I said to my family, why don't we "attend" last week's NBCC? So we did.
Apparently the Mandarin-speaking congregation (in my home church in Sydney) is getting a lot of attendees from Beijing.
When the restrictions were somewhat lifted (so that you could have more than immediate family in your house), we started a system where we rotate around to different people's houses in groups to join the videoconference.
There's no reason we couldn't have done these things before, but we just didn't do it until now.
My brother did CS as well (mid-1980s). There might have been only one female in his year. There might have been more, but he never spoke about them that I can remember.
I suspect the gender bias has been around for a long time.
While dentists can't practice, it's much easier to get time with them to figure out what they need and don't currently have.
The lower the interest rate, the higher the safe level of debt.
Right now federal bonds have a negative interest rate, so perhaps there is no limit to the amount of debt, and in fact that there might be a minimum safe lower bound. Or maybe not; we don't have much experience in dealing with this sort of situation since it was seen to be impossible.
Putting that aside, the UK had debt around 200% of GDP after the Napoleonic Wars, and eventually paid most of it down. So the "bad & unsustainable level" appears to be well above that.
Well, not completely eliminated, but it will be eliminated soon enough based on current trends.