Slashdot Mirror

Finland, where rent control in the private market was gradually abolished in the 1990s? It seems as though currently in Finland, tenants can appeal to the court if they think a rent increase is unreasonable, but otherwise increases should be stipulated in the contract. There are voluntary guidelines but those aren't rent control either. Sweden, on the other hand, has fairly strict national rent control, and has a national shortage of housing while having similar rent-to-income ratios as Finland (second link). In Helsinki, large amounts of land set aside for public housing has contributed to insufficient housing supply, in addition to the long wait lists for said public housing.

Let's also not ignore the fact that Finland's population growth rate is significantly lower than that of San Francisco's, and has many fewer restrictions on constructing new dwellings.

Look, I'm not saying rent control is always bad, just that when it's combined with other policies like it is in SF, it can make the problem worse. Rent control (here we're talking about regulations on rent, not constructing more affordable housing) reduces supply. When you already have high population growth (and that growth is primarily people with higher incomes), a really burdensome and uncertain process to get approval to and then start building, a reduction in supply is the last thing you want. Overregulation or misregulation can be just as bad as underregulation.

The lefties can't possibly take the announcement at face value, they would have to concede that the current administration is doing something right.

Exactly what kind of idiot, regardless of partisanship, takes ANY political announcement at face value?

Frankly, even Ronald Reagan KNEW better, so the fact that you're trying to tell me not to look the horse in the mouth, makes me think you're up to some kind of shenanigans.

Irony, captcha is insulted, so yes, I will tell you I find your insistence of not examining the proposal to be insulting.

94% syntax is definitely good, for a machine learning parser. Now if you were to come to the land of rule-based parsers, 94% is the norm.

Google loves machine learning, and it's easy to see why. That's how they made their whole stack. They have the huge amounts of data to train on, and the hardware to do so. It's so seductive to just throw a mathematical model at huge amounts of data and let it run for a few weeks.

Rule-based systems don't need any data to work with - they just need a computational linguist to spend a year writing down the few thousand rules. But the end result is vastly better, fully debuggable, easily updatable, understandable, and domain independent. That last bit is really important. A system trained for legalese won't work on newspapers, but a rule-based system usually works equally well for all domains.

In 2006, VISL had a rule-based parser doing 96% syntax for Spanish (PDF) - our other parsers are also in that range, and naturally improved since then. Google is hopelessly behind the state of the art.

94% syntax is definitely good, for a machine learning parser. Now if you were to come to the land of rule-based parsers, 94% is the norm.

Google loves machine learning, and it's easy to see why. That's how they made their whole stack. They have the huge amounts of data to train on, and the hardware to do so. It's so seductive to just throw a mathematical model at huge amounts of data and let it run for a few weeks.

Rule-based systems don't need any data to work with - they just need a computational linguist to spend a year writing down the few thousand rules. But the end result is vastly better, fully debuggable, easily updatable, understandable, and domain independent. That last bit is really important. A system trained for legalese won't work on newspapers, but a rule-based system usually works equally well for all domains.

In 2006, VISL had a rule-based parser doing 96% syntax for Spanish (PDF) - our other parsers are also in that range, and naturally improved since then. Google is hopelessly behind the state of the art.

CERN said the evidence is five sigma or so for a particle more or less where the Higgs was expected (or perhaps about halfway between where two competing theories expected it), but some now doubt whether the particle CERN found is actually the Higgs. See this recent reassessment: http://sdu.dk/en/Om_SDU/Fakult...

We have something like that at VISL, but with zero statistical or machine learning or AI aspects.

We instead write a few thousand rules by hand (largest language has 10000 rules) that look at the context - where context is the entire sentence, and possibly previous or next sentences - to figure out what meaning of a word is being used and what it attaches to.

E.g.
Input: "They're looking at writing an AI which can in some sense understand what is being said."
Output: http://dl.dropbox.com/u/62647212/visl-eng.txt , http://dl.dropbox.com/u/62647212/visl-eng.png

This kind of system takes longer to develop and refine, but it also doesn't have any of the statistical problems. 95-99% "understanding" of text? Sure, we can do that. Statistics top out long before, and then have to add in rules to get the last 5-10%. And where statistics require giga- or terabytes of text, rule based systems only require a single example of a valid grammatical construct or word usage.

And how does its syntax differs from NASM and AT&T ?

Intel syntax doesn't feel like it was designed by a sadist.

More seriously, this site link covers some differences. Among the things I like much more about Intel syntax: there's no need to add a ton of visual noise with what-should-be-extraneous $ and % symbols, and things like memory indirection is much easier to learn. Compare "[ebx+ecx*4h-20h]" to "-0x20(%ebx,%ecx,0x4)"; the former almost tells you what it does even if you're not at all familiar with the syntax, the latter definitely doesn't.

The main benefit that AT&T syntax has is that they "hungarian notation" their instructions: movb works on 1 byte, movw on 2 bytes, movl on 4. Most of the time this is extra visual noise (I don't need the 'l' to tell me that 'mov eax, ebx' works on 4 bytes), but it does make memory dereferences more concise. With Intel syntax you'll get a lot of 'dword ptr' stuff lying around to tell how much should be brought in from memory.

Why not just get a degree in a *real* medicine that specializes in soft-tissue and anatomy and then practice that?

Well, that's exactly what it is. The education is called Clinical Biomechanic / Chiropractor or something to that effect, at a regular high-quality university (only Danish, sorry). It's a solid, medical, academic education, no woo-woo stuff at all.
In Scandinavia we don't really have the "Straight chiropractic" variety, which might be the source of our confusion :)
Here chiropractors are trained according to the views on the right of this chart, with emphasis on extensive anatomical knowledge.

Let that be a lesson to you trading n00bs!

Moreover, having a second core doesn't make interrupt handling latency decrease

In fact it may even increase the interrupt latency. The fact that a single CPU system could process a network interrupt faster than a dual CPU system was one of the reasons the Horseshoe cluster was build using single CPU systems.

An OO teaching language called Blue actually did require comments. It did have an interesting version of enumerators.

I have never seen a Debian machine in a production environment. Seriously. Every production Linux machine I see is Red Hat. Netcraft seems to agree too. I rarely see Debian, but I see Red Hat all the time when doing netcraft lookups.
Maybe if by admin you mean, "personal SMTP/DHCP/Firewall server on DSL in my closet" then sure.