jbert

33 ms·

241.
by jbert 15y ago
Ugly thought - if "almost all" accesses are Read-Write consistent, maybe an exception list of addresses could be used? Basically do the 4byte cell approach, and make most of your RAM the "fast" endianness, with some 4byte cells "correct" en
242.
by jbert 15y ago
Thanks for this, it's a great piece of work and very easy to read. Nice code. > "it looks like it is a statistical profiler. When the usual invocation takes less than 1ms, it might get the times wrong" I don't understand this. As long a
243.
by jbert 15y ago
Very nice. Is there any chance of a follow up on your comment in the previous thread: "I'll write a post describing what I did to take boot time from 2.5 minutes to 10 seconds." My interests: - how did you find what to optimise? - what appr
244.
by jbert 15y ago
> If anyone is interested in this stuff, just ask and I'll write a post describing what I did to take boot time from 2.5 minutes to 10 seconds. Yes please. Optimisation war stories are always interesting.
245.
by jbert 15y ago
Awesome, so we'll have the tech to consume the hydrocarbons, but where is the production side? What do we need to be able to hook up a nuclear power plant, a supply of clean water and a big air pump and starting sucking in H2O+CO2 and pipin
246.
by jbert 15y ago
Yes, but the process then becomes biofuel, which also soaks up land area and has other issues (warping world food prices). It would also be nice to be able to capture the energy output of other energy sources, e.g. nuclear in a nice, energy
247.
by jbert 15y ago
What about micro-fuel cell + hydrocarbons? The energy density of gas/petrol is fantastic. Recharge time is quick ("pour some in"). http://www.gizmodo.com.au/2011/12/a-gallon-of-petrol-can-pow... http://www.mtimicrofuelcells.com/ Sure, w
248.
by jbert 15y ago
Large numbers of people have home, food and health care as a basic right (e.g. all inhabitants of the UK and also I assume many other countries). I think the major problems are: - People like nicer food/nicer homes. Some (most?) of this is
249.
by jbert 15y ago
I guess that the fundamental problem is thinking of months and years as numbers (and representing them as such). If they were purely symbolic constants, then the expression "January + 1" is meaningless and would throw an error. So, with hin
250.
by jbert 15y ago
Email per se doesn't have centralised servers. You can run your own. The only registration issues (DNS) are related to addressing (your server needs to be able to find the destination to send the email). Your proposal basically sends all em
251.
by jbert 15y ago
The interesting bit for me was the "everything ok" status page. I wonder if any services run their own monitoring system live into their status page? (and uptime stats). If not, I guess the status page is less urgent than "fixing the proble
252.
by jbert 15y ago
Slashdot (the pioneer in this sort of moderation) has always had a 'reason code' for up and down votes (and meta-moderation to check that the reason codes are good). This allows you to do things like filter results to only high scores, excl
253.
by jbert 15y ago
You're probably right, in that there is a useful role for pure "assign value X" languages. But also bear in mind that simple INI style files can be represented as data structure literals in the language. So you could replace your ini-style
254.
by jbert 15y ago
Fair point. But I'd say that such little languages exhibit feature creep over time. And once the two features you mention (looping + conditionals) are acquired, you've lost all the benefits you mention, since you're now turing complete. Reg
255.
by jbert 15y ago
It seems to me that this is closely related to the idea of a domain specific language - "custom" and "embedded". There is a well-trodden path of "little languages" evolving over time to add features. (e.g. XML build files and templating lan
256.
by jbert 15y ago
Perl style tends to use early return and also has the benefit of being able to emphasise the control flow by using a return with a trailing conditional. I.e. return $user if $found;
257.
by jbert 15y ago
I don't think you can see individual job data, but you (and/or your monitoring daemon) can telnet to the port and issue admin commands. e.g. 'status' will list all job types, with #queued, #running and #workers. http://gearman.org/?id=prot
258.
by jbert 15y ago
All keys in the list need to be examined, since PHP needs to work out if you're updating an existing key or inserting a new one.
259.
by jbert 15y ago
Agree, with the caveat that those who carry the 3am possibility are allowed to delegate at their own risk. i.e. the real rule is "nothing happens to prod without the agreement of those at the sharp end". That allows good devops fun to happe
260.
by jbert 15y ago
By the same argument, nil != true. This could lead to a tri-state ifs (with true/false/nil branches). Or you can just consider that 2-branch 'if' runs the first branch if the test expr is true and the second 'otherwise'. (So that doesn't ma
261.
by jbert 15y ago
'"get-viagra-cheap-now" has tried to contact you, do you wish to allow them' - the spammer has just succeeded in putting their message in front of you. If you had 500 of those per day, it would constitute spam which would again need the sam
262.
by jbert 15y ago
Well, the same problems apply to postal mail, IM and phone calls, etc. I think the two properties which a comms method needs to have a spam problem are: 1) it is inexpensive to communicate with someone 2) it is has a fixed, human-memorable
263.
by jbert 15y ago
Yes, I've used this and it's great for anything where you want a C-stack backtrace. As a variant, if you want a language-level backtrace in an interpreted language, you can: - set up a signal handler which dumps a (perl, python, etc) level
264.
by jbert 15y ago
The reason I asked "slower than what", is because at no point have I claimed that going to memcached would be faster than a local hash (with locking). Whereas I have (in this thread) claimed that an interpreter without a GIL (and with fine
265.
by jbert 15y ago
> What about all situations where you want to use threads, but not due to resource constraints? I would assume your answer would be "don't use threads" but you say you're not rabidly anti-thread. But you can use threads now,with the GIL
266.
by jbert 15y ago
> So you concede Antrix's point that mutlithreading is not slower? Slower than what? I said (and you quoted): threading doesn't come "for free", in that you need to pay a performance cost in terms of locking in your interpreter and/or a
267.
by jbert 15y ago
That's interesting, thanks.
268.
by jbert 15y ago
> That is exactly what you are doing when you decide which of your data structures go in memcache and which don't. Yes, apart from the fact that the ones you don't think about are private. i.e. processes are "private by default" and thr
269.
by jbert 15y ago
> I submit the fact that is nearly 100% is effect, not cause. Well, there's also the point that your shared state needs to be process-external as soon as you want to scale beyond a single process. (e.g. need to run on more than one serv
270.
by jbert 15y ago
> Not everyone shares your narrow requirements when using a general purpose language. IMHO, if the resource difference between processes and threads is significant for your workload then moving away from an interpreted language is proba

More ›