jbert

47 ms·

391.
by jbert 18y ago
> So, doesn't that just mean that a statistically significant portion of the regular HN users would have to respond? No. If 50% of HN visitors voted but 99% of those over 40 didn't (due to whatever confounding factor you like - say emb
392.
by jbert 18y ago
It's effectively worthless because it needs to be qualified so heavily. Dropping the qualification (as will happen sometimes) makes it more like misinformation, which is worse than worthless, so it averages out to zero... "The median poll s
393.
by jbert 18y ago
> Things that are actually more efficient don't require subsidies.... This rings true, but I'm not sure. Isn't the truth the weaker statement that new ventures which can't clearly demonstrate, in advance, that their balance of return+ri
394.
by jbert 18y ago
> I see some cargo-cult potential in fsync-after-every-write here. Well, we currently effectively have cargo-cult no-fsync, in that application writers who don't consider the matter get the fast-but-risky behaviour. I think I'd be happy
395.
by jbert 18y ago
> Fuji was making point-and-shoot cameras with lower megapixels but superior image quality since '05 Could you provide some references for this, please? Not because I disbelieve you, but for the opposite reason. I have an aging low-mega
396.
by jbert 18y ago
OK, but # of shards is fixed "for ever" under the modulus scheme? You pick it once, when you first shard and then you're looking at downtime to adjust it? In order to split across multiple dbs, you're looking at creating say 100/1000 dbs in
397.
by jbert 18y ago
Ah, OK. So you start with enough shards "for ever" and not change the number? Edit: no wait. If you can split a shard across multiple machines, what's the benefit of having more than 1 shard? Why not have 1 shard split across 1000 machines?
398.
by jbert 18y ago
Hi, is there any chance you could have a look at this comment: http://news.ycombinator.com/item?id=497070 and let me know how you resolve this problem in practice? (Or correct my understanding if there is no problem really).
399.
by jbert 18y ago
I don't understand how that scheme can work, since changing the number of shards changes the location of most users. e.g. we have 4 shards, so user 5 is on shard1. If we go to 6 shards, user5 is now on shard5. I guess it works with downtime
400.
by jbert 18y ago
Because a failure in your test code means some application code isn't tested as well as it should be. It doesn't mean the application code will fail. A failure in your application code can be much more serious. Hence, in general, you test y
401.
by jbert 18y ago
That's the nice thing about the "shoulders of giants" though. By and large, scientific progress is cumulative. The cutting edge of quantum theory a few decades back is now undergraduate-level. Yes, there are probably theoretical limits to h
402.
by jbert 18y ago
> However, if the market chooses the testing procedures, it will reward metrics which are harder to game. How would you achieve this? > This would take several iterations (read: many years) to stabilize, but it should converge on so
403.
by jbert 18y ago
> I know there is one glaring problem here More than one, really. State schools in the UK are incentivised in a way similar to this. What it results in is gaming of the performance measures. For example, pupils are steered into softer s
404.
by jbert 18y ago
No love for the binary formats? Let's hear it for ASN.1 ( http://en.wikipedia.org/wiki/ASN.1 ), so good that google can reinvent it as protocol buffers ( http://code.google.com/apis/protocolbuffers ) and get people interested. More seriousl
405.
by jbert 18y ago
Allow people on your aggregator site to manage the karma of the reputation sources. Open to gaming, but all karma is to some extent.
406.
by jbert 18y ago
"eliminate government-owned housing so that these properties can be returned to the tax base and used to their full potential" Would the government really make more ongoing income from taxes on rent rather than rent? That seems unlikely.
407.
by jbert 18y ago
While shouting like that is unlikely, I wonder if some data centres are effected from traffic rumble or other sources of vibration. (Having your server farm performance dip when you switch to diesel generation might be a problem). I guess t
408.
by jbert 18y ago
Interesting stuff, but I'm surprised by the observation that number of children correlates positively with income. I would have guessed the opposite. Does anyone have the data?
409.
by jbert 18y ago
"If you give away research time for free, or you chase a dead end and decide not to bill it, do you set a precedent that those things will be free in the future?" I don't think so. You set a precedent that you will be fair to the customer.
410.
by jbert 18y ago
As far as I know, you'll get no answers here (or anywhere). My view of things was shaped a lot by some sections of Hofstadter's Godel, Escher, Bach. If you allow that symbols (think of a bunch of object properties + some methods) can be rep
411.
by jbert 18y ago
Nice link and info, thanks for that.
412.
by jbert 18y ago
If large organisations are less efficient: - why do they survive? - if they survive despite all, why do we permit them? If the market was regulated to set a cap on the size of a single company's workforce (or some other metric, perhaps mark
413.
by jbert 18y ago
He seems less legacy obsessed to me than Tony Blair. Large chunks of the UK media have had it in for him following some initial mis-steps (the will he/won't he election-call thing comes to mind). As far as I can tell, their story then becam
414.
by jbert 18y ago
"I'm just trying to nail down precisely what you are trying to say." He's trying to say he doesn't like Gordon Brown. re: your original point, I think in this context, by 'Democracy' people mean "honest politicians working towards ends they
415.
by jbert 18y ago
In the perl world, I: - use apt to manage what's installed on the system - use CPAN to manage additional stuff installed per-user (you just specify a standard install prefix for all CPAN modules) that way I don't mess with the system (any p
416.
by jbert 18y ago
We enhance open source software for use on our platform. We'd love to get our changes into upstream, but it isn't always easy to make the time for that. As you say, all kinds of issues are raised by having our own variant, but fundamentally
417.
by jbert 18y ago
I think the argument is that google is an order of magnitude or so better at downtime than you are, so any google-created downtime is lost in the noise.
418.
by jbert 18y ago
To me, one of the more interesting aspects of that was that the ruby version optimised for reduced memory usage got a 2x speedup. That makes sense in terms of cpu cache misses. Has anyone run any numbers to see how gcc -Ospace compares with
419.
by jbert 18y ago
OK, thanks. So the problem is that transistor density is going up faster than pin density? Makes sense. I guess one approach to that is a bigger die size, so you end up with a fixed transistor/cm^2? That would mean die sizes doubling with M
420.
by jbert 18y ago
Negative, sorry for abbreviation.

More ›