mjb

61 ms·

511.
by mjb 14y ago
This kind of little speaker-microphone 'sonar' is the kind of project we would give to undergraduate students as an afternoon practical. It's simple stuff to get going. The real magic that would make this useful is the signal processing and
512.
Pat Helland - Idempotence Is Not a Medical Condition(queue.acm.org)
2 points by mjb 15y ago|1 comments
513.
The Five-Minute Rule Ten Years Later (1997)(research.microsoft.com)
4 points by mjb 15y ago|0 comments
514.
by mjb 15y ago
> The cache dropping is actually useful when you are doing benchmarking... Agreed. My response was more to the "Let's Put 'echo 3 > /proc/sys/vm/drop_caches' In Cron and Get Free RAM!!!!" thinking, which sadly seems to be widespread
515.
by mjb 15y ago
Yes it can, but you probably don't want to do that. You definitely don't want to do it in an automated way when the machine is experiencing memory pressure. There are very good reasons that Linux (and most other modern operating systems) ma
516.
by mjb 15y ago
> This touches on what I think is one of the more interesting technological problems that we have to worry about right now. As a disclosure, most of the reason I think that is because I'm a PhD student doing research on hardware securit
517.
by mjb 15y ago
I agree with you - this article does a very poor job of communicating the advantages of 'NoSQL' databases over traditional solutions. It spends too many words setting up a straw man (SQL == not partitioned, not highly available, not redunda
518.
The Bleak Future of NAND Flash Memory(cseweb.ucsd.edu)
2 points by mjb 15y ago|0 comments
519.
by mjb 15y ago
One issue of concern is the possible cost of these false positives. If doctors are very careful to communicate to patients that 75% of positive results will only require a second test, then that's OK. Sadly, that doesn't seem to be what hap
520.
by mjb 15y ago
To me, this is a great argument in favor of building on top of products like Akka rather than implementing the same functionality yourself. It's pretty clear from Doug Lea's description ( http://cs.oswego.edu/pipermail/concurrency-interest/
521.
Two different conclusions about drive failures and temperature(brooker.co.za)
2 points by mjb 15y ago|0 comments
522.
David Patterson - Latency lags bandwidth (2004)(ll.mit.edu)
2 points by mjb 15y ago|0 comments
523.
by mjb 15y ago
> My experience is that most 1.5+ developers... My experience is that most of the highly skilled engineers that I have worked with don't really care about the language that much. Most of them are much more interested in what they are bu
524.
by mjb 15y ago
I was mostly agreeing with you until you starting on the technology bigotry. > Because while the Java+IDE environment makes it extremely difficult for an individual engineer to have a 1.5+ impact –there are two cases in recorded history
525.
by mjb 15y ago
The answer to most of your questions is 'maybe', unfortunately. Reasoning about the things that could happen when memory errors occur is very difficult, because they occur outside the mental model of computation that most programmers (and s
526.
by mjb 15y ago
Why is this being downvoted? The processor used (i-7 2600) does not support ECC RAM, and this platform does not include it. A server platform without ECC RAM is a pretty big disadvantage for several reasons, to most compelling being the hig
527.
by mjb 15y ago
> you can't write ... and get something that works. That's taking things a bit far, isn't it? Sure, you can't exactly express 0.1 as a double - but you can get many things that work really well using the approximation.
528.
by mjb 15y ago
> I don't know how to estimate what values of n would be relevant for today's applications, but n=2^22≈4M doesn't seem like such a big number for images or video (if signal length roughly corresponds to number of pixels). Any thoughts?
529.
by mjb 15y ago
Spare transforms like this are very interesting, and these seem like some very good performance results. It's worth noting, though, that FFTW ( http://www.fftw.org/ ) is still much faster on the sizes of transforms that many (most?) applica
530.
by mjb 15y ago
> In other words for a particular millisecond seq order is only relevant on that particular node, so node should come first. It still seems to me that T_0,0,A (issued by node A at T_0 with seq number 0) would tend to come before T_0,1,B
531.
by mjb 15y ago
A couple of questions about this: - Why [time, node id, seq] and not [time, seq, node id]? That would improve ordering if you have approximately equal load on each box. - Isn't a 16 bit seq number overkill? Handing some of those bits to the
532.
by mjb 15y ago
I'm not sure about this one. On the one hand relevant search results and less spam are good things for me. On the other hand, one of the key things I have always enjoyed about the Web is being exposed to new people, new ideas and things tha
533.
by mjb 15y ago
> Test cases were not that fashionable back in the day. Surprisingly, people still got work done. Reinforced concrete did not exist back in the day. Surprisingly, some people still got work done. It's tempting to be dismissive of modern
534.
by mjb 15y ago
The authority for hackerne.ws is domaincontrol (GoDaddy, apparently) while the authority for news.ycombinator.com is EasyDNS.
535.
by mjb 15y ago
As is mentioned in the article, using cryptographically secure hash functions doesn't help - unless you also provide some salt or otherwise randomize the function in some way. On the other hand, if you are already randomizing the function i
536.
by mjb 15y ago
Be careful with load testing. When your business/community/etc is 10x the size it is today, it's unlikely that usage patterns and use cases will be the same. Rather, they are likely to shift as you grow. Simply firing 10x today's load at th
537.
by mjb 15y ago
> This is nontrivial, but is much easier than scaling your own custom stateful code. Yes, this is an importance piece - and disagrees somewhat with your "don't think about scaling yet" point. Probably the most important part of designin
538.
by mjb 15y ago
> While I tend to agree that modeling and approximation error are usually greater than floating point error, floating point error can be more devastating in practice because once you realize you have a floating point problem, it's much
539.
by mjb 15y ago
> This is even difficult to do when what you're trying to model is not the physical world, but a computer system. Indeed. A couple of days ago, somebody posted a fascinating image of memory latency ( http://www.bitmover.com/mem_lat.jpg
540.
Clayton M. Christensen - How will you measure your life?(hbr.org)
3 points by mjb 15y ago|0 comments

More ›