I'm sorry if I sounded like a jerk, that wasn't my intention.
can you explain how a cookie dropped on my browser by abc.com will be used to correlate my browsing on the site xyz.com? Of course, they have only scripts and resources originating from their own domain, which is the assumption in this thread.
The HTML rendered by xyz.com includes a script tag resulting in a fetch of abc.com/foo.js. That HTTP request to abc.com/foo.js comes not only with the cookie which was set last time you fetched any other site using abc.com's foo.js, identifying who you are, but also an HTTP Referer header indicating which HTML document it's being loaded from, indentifying what you're looking at.
They may share all the data in the world, they won't know how to match it with the other domain data. Google cannot do its job with analytics even if I forward all the requests server side to them. The cookie they dropped on xyz.com won't show up on my browsing data. They won't be able to correlate.
False, I'm afraid
Third party cookies: It should be only the cookies from the page you see the URL in the browser address bar that are allowed. None other.
Some-origin policy applies to the URL of the script js, not the document using that script.
Of course it is, for the simple fact that the power a battery can deliver scales linearly with its capacity.
Yikes; that is wrong on several levels. Battery capacity (quantify of work; watt-hours or nom. V * amp-hours) has nothing to do with the power (rate of work; watts or nom. V * amps) it can deliver. For example, you can find many batteries sharing the same nominal voltage and watt-hour capacity, but very different maximum discharge current (expressible as watts). Batteries are optimized for the application to which they are intended; the ideal 300 watt-hour UPS battery bank which needs to support 2000 watt loads for a few minutes will be made of very different batteries to the ideal 300 watt-hour solar-charged, remote SCADA monitoring station which only draws a few watts for days at a time. The batteries in our solar application would melt if you tried to put them to work in the UPS, even though they have the same watt-hour capacity.
1 Wh = 1 J
1 Wh DOES NOT EQUAL 1W (which is more or less what you've been asserting, over and over)
no I haven't
What I've said is,
if 1W = 1J
then
1Wh = 3600J
This is wrong, so continuing this line of thought doesn't seem useful at all.
The "h" is entirely a red herring, I've consistently pointed out it has no meaning here.
What I specifically said is
Rossi is saying he can deliver 1kW of heating for $150
and this is non competitive with conventional fuels which can deliver 1kW of heating for a few cents.
There is no way for us to convert this into kWh, other than to guess he means "1kW for a short period of time".
No need to guess; usually, if an electricity generation system is rated at 1kW, it should support a 1kW load for as long as its energy source holds up, "for a short period of time" or otherwise. A 50kW generator says nothing about its efficiency at turning diesel into electricity, or the size of the diesel tank it's attached to, or even how long that diesel would last for (how big is the load?).
TFA uses language such as
We must make a distinction between the price of the industrial plants and the price of the domestic plants.
- and in that context, a unit of kW rather than kWh does make sense; although if he really could build 1MW for $150k, regardless of the energy source, I'm sure he'd be overwhelmed with customers.
Quite bluntly, the 3kW here says nothing about how long the UPS can hold up for, with a 3,000 watt load or a 1 watt load.
Actually, that isn't true is it.
You seemed to be mis-using maximum watt ratings in your discussion, and invited a UPS example. So the point I am making is that, in a discussion trying to ascertain the amount of useful work which can be extracted from a given generator+energy storage system, the maximum load rating of the generator component (or the energy storage, for that matter!) is not helpful.
Eg., for a UPS system, the KVA rating has nothing to do with the amount of useful work it will provide (aka "how long will run for"), other than the fact that manufacturers tend to sensibly size the internal batteries (if supplied) to roughly match a nominal utilisation (typical load scenario) relative to the advertised maximum load rating of the inverter circuitry.
So, UPS runtime has everything to do with load (watts) and battery capacity (watt-hours). We do not care about UPS (or battery) maximum ratings, unless the load under consideration cannot be supported, or cannot be supported efficiently.
Well, we could also talk about battery efficiency, inverter efficiency, power factor of AC loads, etc. but again, nothing to do with maximum ratings.
Let's say you've got your favourite 3kW UPS in the rack. Quite bluntly, the 3kW here says nothing about how long the UPS can hold up for, with a 3,000 watt load or a 1 watt load. This isn't a simple number - it depends on the LOAD (measured in watts) and the CAPACITY of the batteries (usually spec'd in amp-hours, but could just as well be watt-hours, average discharge voltage x 1hr discharge current (amps) ~ watt-hours).
So a 3kW UPS just means it can prop up a 3,000 watt load before it buzzes with the front panel screaming "O/LOAD".
But you didn't cheap-out buying the smallest UPS you could get away with, you only have 1kW of stuff attached. Let's pretend 220V gear (and ignore RMS/Vpp/duty-cycle/switching noise/efficiency madness). That means we've got 1000/220 = 4.54 amps of current flowing through the output cables and into your gear.
But say we've replaced those four little 12V DC batteries inside a few times, and we notice they're rated at 7.2Ah (amp-hours). That means those batteries all up are storing 12 * 7.2 * 4 = 345.6 watt-hours of energy (assuming the batteries provide an unwavering 12V AND 7.2 amps continuous output over the entire discharge cycle... yeah right!) AND the UPS is 92% efficient converting the battery power into 220VAC (and the AC waveform is a perfect sinewave, among other lies) into our 1,000 watt load
That's 1000 / 0.92 / 12 / 4 = 22.64 amps flowing through those big fat battery cables on the input side.
Which means when our ideal-batteries suddenly stop working, they will have provided their 345.6Wh after 345.6/1000*0.92*60 = 19.07 minutes. Or, calculate it from the battery's 7.2Ah rating: 22.64 amps from an 7.2Ah battery bank means exhaustion in roughly 7.2/22.64 = 19.08 minutes.
+1. We use Foswiki, with its DataForms for consistently structured data, SemanticLinksPlugin for ad-hoc relationships (though relationships themselves are also managed, sometimes with more SMW-style links), JQGridPlugin, MongoDBPlugin to ensure our poor VM can keep query speeds up regardless of how many pages we get, DirectedGraphPlugin (graphviz) for simple network visualisations and an http://thejit.org/ based interactive js visualisation to explore our entirely wiki-based datasets, along with an experimental OpenLayersPlugin to plot linked data which has long/lat (biological specimen) info.
IMHO the querying capabilities of Foswiki are far stronger than SMW, which I've not really found any way to do indirect queries with, Eg. the somewhat contrived example: Author/Country/Population < 20000000 (where an Author field connects to a topic which has a Country field that connects to a topic with a Population field holding a number less than 20,000,000) is a valid QuerySearch expression in Foswiki, but maybe SMW has a way of doing similar stuff these days.
Disclaimer: I wrote the (still not-out-of-beta) SemanticLinksPlugin to support SMW syntax for ad-hoc data (vs structured DataForms, think table schema):-)
It is. You walk into the polling place and vote for someone else.
Excellent idea! But I think that's what people thought they were doing in 2007 when we ditched the coalition. I thought the two parties were pretty hard to distinguish in the last election; but I told myself I was just being cynical... now, it's plainly obvious that current politics involves this sick competition between the two parties to be as utterly, completely, and hopelessly exactly the same as each other.
I've never voted for either of the two major parties before (and yes, I fill out all 60-odd boxes so that labor & liberal go last in my preferences), having said that, I'm an extremely reluctant Greens voter. Most of their underlings seem incompetent, or are just as warped as any other career-politician, or they remind me of people I met at university who seem capable of instantly forming very loud, unshakable, annoying opinions on any subject regardless of their understanding, knowledge, and painfully obvious lack of research.
How does one establish whether methods/vars are public/private/protected? Or inheritance? To me, the weird misappropriation of the function keyword to build objects, the verbosity of the code to express objects, and the lack of inheritance, etc. are primitive compared to Actionscript 3, to Java, to PHP5, to C++, and a variety of other languages I've dealt with.
You really need to read Crockford's "Javascript: the good parts". You absolutely do make private methods and vars (ever noticed that you can't directly call jQuery's internal methods? Or TinyMCE's? Or any other major library/framework?)
He also makes the case that actually JS has more patterns to allow code re-use. That's why things like Mootools can even fake things that look like classical class inheritance patterns for you, if you really want to do that.
Whereas trains use HIGH voltage and current AC.
When building a capacitor, you have to fight against conflicting requirements: high density, high current, high voltage, stable against environmental changes (humidity/temperature), stable against aging characteristics, stable against voltage/charge (and voltage/charge rate), AC/RF response characteristics, dissipation ("leakage"), among other things.
Double-layer capacitors ("ultra-caps") sacrifice maximum operating voltage and maximum discharge rate (current) for charge capacity/density.
You can use SQL for everything, or you can use the best tool for the job. Usually that's SQL, but the other choices aren't just for big data: graphdbs like neo4j allow you to efficiently query deeply, arbitrarily linked data in a way that just can't be done with SQL.
And then there's the triple/quad-stores like jena, 4store, virtuoso which allow you to answer questions from your data that you couldn't even begin to imagine doing with SQL.
So although I think these graph & "triple" DBs are the way of the future, it's an extremely crazy thing to ditch SQL for your core business data. But it's also crazy to dismiss these non-SQL technologies simply because they aren't SQL - there is a lot of potential in having your data in one or more of these "alternatives", similar to the value you get from wiring up solr/lucene for full-text search.
We've been doing a project with MongoDB (no, we're not using it as our authoritative data store, but neither are we using SQL for that), which we could have done with an SQL DB, but honestly we didn't have the resources to do so. We have less than a million records, hardly a huge amount of data - our data is arbitrarily structured, fitting the document-object model perfectly, and makes for absolutely meaningless SQL tables (with many joins, stored procedures so that we could try to service an obscure perlish query language).
Additionally, the ability for us to delegate some parts of the more complex queries to JS on the server(s) is incredibly useful. And the integration cost was very low: the perl driver is officially supported by 10gen, it fit quite naturally into the project... the proof of concept was done in just a couple of weeks.
Our contractor has, however, discovered some irony with MongoDB's "schemaless" claim: unless you can pick an extremely finite set of fields to be indexed, you can't actually do arbitrary queries on arbitrary documents... in other words you need to decide on a schema:-)
Thanks for the feedback. wl txpwr1 shows 28mW, as I set it from rc.common.
I know that the APs are broadcasting way too strongly by default with 7.09. At one site I was approached by one of the users asking why he could get a reasonable signal strength but not connect. He was trying to access it from a location that I know was previously unusable under White Russian. A few days later, even people close to the AP couldn't reliably stay connected.
By the time I returned to the site a week later, none of the APs are able to transmit at all. A site survey for instance, finds other APs, but no laptop, other AP, or my nifty wlan meter thing can see any trace of a signal whatsoever. Try reflashing back to Linksys firmware, no luck. Good as junk.
Had the same problem at the other site. These had more time to kill themselves though, as the network was unused for a few months (inter-office LAN connectivity project that was never utilised). I replaced cables and antennas until I discovered the APs themselves weren't transmitting (but were able to scan and report other networks via site survey). Rooftop/weatherproofed units too, not easy to replace.
It seems bizarre I'm the only one suffering this problem. I'm running 12V PoE over 20m of Cat-5e. Perhaps the DC-DC conversion after a 2 or 3v drop (~10V at the device) induces some kind of noise that stresses the transmitter circuitry somehow.
Who knows. I do know that I don't want my APs defaulting to 127mW transmit power.
It still would have been handy to check/confirm transmission power setting from what is otherwise a fantastic web interface on 7.09. It's simply impossible without installing the wl package, and this oversight surprises me.
My confidence in the platform was never all that high, but people are wary of spending the money on what I was using previously (Nexus Airpoints).
I just set another WRT54GL up today, it reports 127mW. What does yours say?
I'm not sure what I'm doing different, but it always defaults to maximum without running wl at boot to set it otherwise.
The rest of Kamikaze is brilliant, which is why I persist with it. I've compiled my own firmware now so that wl and the modified init script is there by default (along with OpenVPN, knockd, QoS, etc).
I'm running WRT54GLs, v1.1, all from the same supplier. The unit I just configured has S/N: CL7B1H416073.
It takes about 3-7 days for the AP to finish killing itself. I installed four at one site, and two at another, then forgot about them as I usually do. User complaints were not immediate.
It's not the first time I've heard this, but it comes from people who've observed degraded performance after increasing their AP's power output (usually with a 3rd-party firmware).
What's going on here is that:
a) Clients are still transmitting at normal power, so the AP can't hear the clients.
b) Many APs are built with circuitry that doesn't like to be pushed very far beyond factory defaults with transmit power: the signal really does get "noisy" at high power settings.
Too bad the default Kamikaze 7.09 OpenWRT firmwares kills any and all (six!) WRT54GL routers that I put it on (previously ran White Russian brilliantly). Apparently flashing these things with said firmware out of the box defaults the output power to 150mW (default is 28mW), and fries the transmitter circuitry. There's no option to fix this, you're supposed to install a package onto the router called "wl" and hack a call to this utility in the init script for yourself that sets the output power at bootup.
Actually, wifi is more like an old-school hub (vs switch) than you think: all clients talk and listen on the same medium (radio vs wires) at the same time, the only difference being that wifi networks have the added problem of not being able to tell if they've caused a collision while transmitting.
To successfully share the bits around, wired hubs employ CSMA/CD, whereas wireless does CSMA/CA - Carrier Sense Multiple Access, Collision Detection vs Collision Avoidance.
As the number of communicating nodes increase, you get pretty much the same (drastic) drop in the total amount of data shifted (due to collisions/contention) as you did with wires (using a hub).
I've also mostly developed in C (on embedded systems), among other languages (Verilog on FPGAs, for example).
At work, I chose Ruby (could have chosen Python, but found Ruby more expressive).
I would have considered Java more seriously if the company I worked had more than one development engineer (me) and had a more rigorous engineering style (at the moment, virtually none).
So I'm sort of hacking on the fly. It's a large app, for me, possibly the largest scope for anything I've undertaken before.
But I could not have done it in C/C++/Java - customers don't know what they want, so much pie in the sky features they desire. With Ruby I've been able to meet deadlines and even implement their crazy ideas without as much pain as I would have had with Java.
So there you go. Java can thrive in properly engineered, bureaucratic apps with the resources to support that deveopment style.
Ruby thrives in a more ad-hoc, rapid development. You may even consider it a prototyping language (in fact, I have done that once: knocked up a prototype web interface for an embedded project for the customer to confirm what they wanted. Then it took two months to do the same thing in C).
If you can't see the merits (and pitfalls) of where dynamic languages are useful, you need to play with them more;-)
I've been tasked with coming up with a GUI that involves data visualisation and report presentation, where before I've mostly done very discrete back-end or embedded systems stuff.
Because there's real-time data visualisation (as well as historic stuff), I've heard about the Tufte books before and so bought all four available at bookware.com.au - Beautiful Evidence, Visual Display of Quantitative Information, Envisioning Information, and Visual Explanations.
Still waiting on them, probably won't be able to appreciate them all in time, but hopefully I can make my app suck less than the existing solutions I'm tasked to replace.
My application is loosely what might be traditionally known as SCADA... but for various reasons we're not using traditional SCADA packages. We're presenting industrial process data, traditionally there are real-time figures and "dials"/bar graph gauge type indicators, along with graph plots that resemble the paper and pen chart recorders this software replaced many years ago.
Any particular one of the four books that people might know to be most useful for me, or a suggested reading order anyone might have?
Because the statement that it would be equivalent to "The power output of the Sun for.39 nanoseconds" is misleading.
Don't get distracted by the 39ns figure. Power is an instantaneous quantity - it is a rate at which energy is transmitted. They are saying that the bomb sustained a level of power (rate of energy) output and held it there for a period of time - 39 ns - that approached 1% of the sun.
I repeat: 39ns is just the period of time that the power level peaked for. They calculated that the amplitude of the power peak itself, was equivalent to 1% the power output of the sun.
We don't care about how long the peak lasted for, the 39ns, unless you start integrating power over time as you just did, in which case you're comparing a quantity of energy, rather than a rate of energy output. Yes, I suppose you could say that 39ns @ 1% sun power is equivalent to an amount of energy produced by the sun in 0.39ns, but that's not the interesting number here, because we could similarly integrate just about any huge power source over a long enough interval of time (hours, days, years, whatever) to come up with "the same amount of energy output by the sun over 39 ns".
So the interesting number is in this case, yes, that the actual instantaneous absolute power output of the bomb approached 1% of that of the sun, albeit for only 39 ns.
Wow, you are being very picky today. Saying that "rising voltage does not blow fuses, rising current does" is irrelevant... There was nothing technically wrong with parent's observation on the behaviour of protection circuits when presented with excessive voltage. Who knows, a transient voltage suppressor may be protecting the load as well as that fuse, and the TVS may provide that current to make the fuse blow on over-voltage.
My background is admittedly low power engineering, albeit in hazardous areas (intrinsically safe systems where there's explosive gases in the air). As for utility power transmission lines, I'm pretty sure those fuses are usually designed so they don't spray shrapnel everywhere. Particularly at substations the line fuses are quite exotic, composite fuse links (involving silicon), immersed in spark-arresting gases and so on. On the LV side of some plants I've seen them use two different types of fuses in series - one to interrupt "dead-short" type over-current, and the other for more gentle 200% overload situations.
But all this talk of how fuses work is... redundant? These cabinets are using unstable batteries that do require precautions to maintain a safe working life. I can tell you that in IEC Zone 1 and Zone 2 hazardous areas (these might exist at mine sites, chemical processing plants, petroleum refineries, etc), it's very difficult to certify lithium batteries in this environment - they are just not trusted. A certified Lithium battery has electronics to protect each *cell* of the battery: monitoring internal pressure of the cell, temperature, preventing discharge below a safe voltage, preventing overcharge to above a safe voltage, and of course solid-state current limiting, as well as fused protection. And did I mention that's on each cell inside the battery? Because all it takes is one faulty cell to really, really ruin your day.
I don't understand what there is to discuss... unless someone blew it up with explosives, the unstable batteries in the compartment lacked sufficient internal protection circuitry to prevent explosion. Maybe a faulty charge circuit coupled with a dud thermal fuse in the battery (or all the batteries, if there was a faulty batch) caused all the cells to heat up, internal pressure exceeded casing maximum, it vented flammable material which ignited, starting a chain reaction by igniting the rest of the cells/batteries that were already just about boiling and ready to pop.
The fact that everyone acknowledges here is that Office is the defacto standard for document markup.
"Document Markup" is an interesting way of describing of.doc and friends.
The ISO process does not require standards to be open.
And yet, Microsoft prance around with the "Open" prefix. And yet, their RAND patent license excludes free software.
Meanwhile Sun's proposal is just as proprietary as Microsoft's, neither is the process of an open design process, they are merely a schema dump from an existing program.
The difference being that Microsoft's spec has things like "do it the way Office 97 does it", and the ODF spec doesn't.
The simple fact that there are other Office suites already reading and writing ODF files other than OOo/StarOffice (Abiword, KOffice for example) demonstrates that it is a viable and workable standard.
It's my impression (others have read more of the 6,000 pages of documents than I have) that the same could not be successfully achieved from the OOXML spec.
And Sun has a vastly worse history as far as open standards go, suing companies for not implementing Java in their prefered maner.
That's funny, exactly what Microsoft seems to be planning. Their royalty free patent license may only be granted if you implement their standard EXACTLY (a herculean feat in itself). Want to enhance or modify your software, as the GPL explicitly sates you should be allowed to do? Sorry, you just agreed to get sued by Microsoft..
They weren't doing so before? I don't know about the US, I presume it's fairly similar to AUS. We call them student unions, is it really any surprise, that students would organise in such a manner to look out for their interests?
Actually, at my uni, having anything to do with the student union was a novelty that a very small minority engaged in first year, and after that... well, there isn't a single person in my engineering year (or below) who I know of that gives a damn. In fact, the student in my year who "won" the seat to represent the students within the engineering school was sort of forced upon them.
Now with VSU (voluntary student union), the student union is even more pathetic than it ever was.
Not that I really mind - my union fees (easily as much as 3 text books) should be spent on student services at the campuses, not on rudderless effigy burning protests that in the course of a few hours behind the megaphone apparently end up with 15 different agendas, introduced whimsically as if to keep themselves from getting bored, ranging from gay rights for albino homing pigeons to implying that our welfare system for full-time unemployed students is somehow below "the poverty line" (granted, austudy + rent assistance isn't decadence living, but I hate ignorant people using hyperbole like this because it trivialises the real issue they're using as an attention getter - and the reality is these idiots have never come close to trying to survive actual "poverty line" conditions as defined by the UN. For instance, I don't see any of the African or Indian students who have come over from very poor backgrounds on a life-changing scholarship programme, joining in their protest and agreeing with their cause).
Then there's the bus trips for protests out at woomera and embezzling e.g. print costs for the student magazine out to a mate's print shop instead of using much cheaper university print facilities... oh, and doing their damn job that some of the union positions were paid to do, instead of sitting around caring more about influencing "the system" (in any way possible) than the actual issues they use as a platform to accomplish this.
In principle, I know that unions, and student unions, are very useful when they're needed. But when they exist for the sake of existing, they invent their own causes to justify their existence, and somehow it breeds a culture of people who are in it for the "scene" more than anything else... Creating an attitude that alienates them from the vast majority of the student body who have more important things to worry about.
The issue, as far as I can gather, is that an attorney charged with protecting the electoral system against fraud, was actually working for an office which itself is challenging and invalidating millions of otherwise legal votes.
The E-mails gave specificity as to how this was done. I'm under the impression the sheer scale of vote invalidation should be setting off alarm bells, but maybe it's a normal number. Regardless, IIRC they have 70,000 names on the "caged" list. I suppose it could be an elaborate hoax to frame the Republicans - which seems to be what you are suggesting could be a possibility?
The company that received the E-mails erroneously is some anti-republican group, clearly - georgewbush.org is dripping with sarcasm.
There appear to be a lot of very high-profile people visibly upset at these E-mails. I haven't heard any suggestions that imply they could be fakes.
Either way - the sheer number of votes challenged, along with the rest of the circumstantial stuff (like, the president of the USA's entire E-mail system losing every E-mail the subpoena was interested in) makes me think the ball is more in the Republican's court to explain and defend their actions.
NB: I'm not from the USA. And I don't like the sounds of the democrats either, not that I'm an expert at having an opinion on US politics. Just sharing my view so far from what I can see.
Slashdot Mirror
I'm sorry if I sounded like a jerk, that wasn't my intention.
The HTML rendered by xyz.com includes a script tag resulting in a fetch of abc.com/foo.js. That HTTP request to abc.com/foo.js comes not only with the cookie which was set last time you fetched any other site using abc.com's foo.js, identifying who you are, but also an HTTP Referer header indicating which HTML document it's being loaded from, indentifying what you're looking at.
True, but...
False, I'm afraid
Some-origin policy applies to the URL of the script js, not the document using that script.
Of course it is, for the simple fact that the power a battery can deliver scales linearly with its capacity.
Yikes; that is wrong on several levels. Battery capacity (quantify of work; watt-hours or nom. V * amp-hours) has nothing to do with the power (rate of work; watts or nom. V * amps) it can deliver. For example, you can find many batteries sharing the same nominal voltage and watt-hour capacity, but very different maximum discharge current (expressible as watts). Batteries are optimized for the application to which they are intended; the ideal 300 watt-hour UPS battery bank which needs to support 2000 watt loads for a few minutes will be made of very different batteries to the ideal 300 watt-hour solar-charged, remote SCADA monitoring station which only draws a few watts for days at a time. The batteries in our solar application would melt if you tried to put them to work in the UPS, even though they have the same watt-hour capacity.
no I haven't
What I've said is, if 1W = 1J then 1Wh = 3600J
This is wrong, so continuing this line of thought doesn't seem useful at all.
The "h" is entirely a red herring, I've consistently pointed out it has no meaning here.
What I specifically said is Rossi is saying he can deliver 1kW of heating for $150 and this is non competitive with conventional fuels which can deliver 1kW of heating for a few cents.
There is no way for us to convert this into kWh, other than to guess he means "1kW for a short period of time".
No need to guess; usually, if an electricity generation system is rated at 1kW, it should support a 1kW load for as long as its energy source holds up, "for a short period of time" or otherwise. A 50kW generator says nothing about its efficiency at turning diesel into electricity, or the size of the diesel tank it's attached to, or even how long that diesel would last for (how big is the load?).
TFA uses language such as
We must make a distinction between the price of the industrial plants and the price of the domestic plants.
- and in that context, a unit of kW rather than kWh does make sense; although if he really could build 1MW for $150k, regardless of the energy source, I'm sure he'd be overwhelmed with customers.
You seemed to be mis-using maximum watt ratings in your discussion, and invited a UPS example. So the point I am making is that, in a discussion trying to ascertain the amount of useful work which can be extracted from a given generator+energy storage system, the maximum load rating of the generator component (or the energy storage, for that matter!) is not helpful.
Eg., for a UPS system, the KVA rating has nothing to do with the amount of useful work it will provide (aka "how long will run for"), other than the fact that manufacturers tend to sensibly size the internal batteries (if supplied) to roughly match a nominal utilisation (typical load scenario) relative to the advertised maximum load rating of the inverter circuitry.
So, UPS runtime has everything to do with load (watts) and battery capacity (watt-hours). We do not care about UPS (or battery) maximum ratings, unless the load under consideration cannot be supported, or cannot be supported efficiently.
Well, we could also talk about battery efficiency, inverter efficiency, power factor of AC loads, etc. but again, nothing to do with maximum ratings.
Let's say you've got your favourite 3kW UPS in the rack. Quite bluntly, the 3kW here says nothing about how long the UPS can hold up for, with a 3,000 watt load or a 1 watt load. This isn't a simple number - it depends on the LOAD (measured in watts) and the CAPACITY of the batteries (usually spec'd in amp-hours, but could just as well be watt-hours, average discharge voltage x 1hr discharge current (amps) ~ watt-hours).
So a 3kW UPS just means it can prop up a 3,000 watt load before it buzzes with the front panel screaming "O/LOAD".
But you didn't cheap-out buying the smallest UPS you could get away with, you only have 1kW of stuff attached. Let's pretend 220V gear (and ignore RMS/Vpp/duty-cycle/switching noise/efficiency madness). That means we've got 1000/220 = 4.54 amps of current flowing through the output cables and into your gear.
But say we've replaced those four little 12V DC batteries inside a few times, and we notice they're rated at 7.2Ah (amp-hours). That means those batteries all up are storing 12 * 7.2 * 4 = 345.6 watt-hours of energy (assuming the batteries provide an unwavering 12V AND 7.2 amps continuous output over the entire discharge cycle... yeah right!) AND the UPS is 92% efficient converting the battery power into 220VAC (and the AC waveform is a perfect sinewave, among other lies) into our 1,000 watt load
That's 1000 / 0.92 / 12 / 4 = 22.64 amps flowing through those big fat battery cables on the input side.
Which means when our ideal-batteries suddenly stop working, they will have provided their 345.6Wh after 345.6/1000*0.92*60 = 19.07 minutes. Or, calculate it from the battery's 7.2Ah rating: 22.64 amps from an 7.2Ah battery bank means exhaustion in roughly 7.2/22.64 = 19.08 minutes.
So perhaps you're responding to:
I.e. that there's no such thin as 'regular' economics, but then that nit-pick sounds like nerd bullshit too
+1. We use Foswiki, with its DataForms for consistently structured data, SemanticLinksPlugin for ad-hoc relationships (though relationships themselves are also managed, sometimes with more SMW-style links), JQGridPlugin, MongoDBPlugin to ensure our poor VM can keep query speeds up regardless of how many pages we get, DirectedGraphPlugin (graphviz) for simple network visualisations and an http://thejit.org/ based interactive js visualisation to explore our entirely wiki-based datasets, along with an experimental OpenLayersPlugin to plot linked data which has long/lat (biological specimen) info.
IMHO the querying capabilities of Foswiki are far stronger than SMW, which I've not really found any way to do indirect queries with, Eg. the somewhat contrived example: Author/Country/Population < 20000000 (where an Author field connects to a topic which has a Country field that connects to a topic with a Population field holding a number less than 20,000,000) is a valid QuerySearch expression in Foswiki, but maybe SMW has a way of doing similar stuff these days.
Disclaimer: I wrote the (still not-out-of-beta) SemanticLinksPlugin to support SMW syntax for ad-hoc data (vs structured DataForms, think table schema) :-)
Excellent idea! But I think that's what people thought they were doing in 2007 when we ditched the coalition. I thought the two parties were pretty hard to distinguish in the last election; but I told myself I was just being cynical... now, it's plainly obvious that current politics involves this sick competition between the two parties to be as utterly, completely, and hopelessly exactly the same as each other.
I've never voted for either of the two major parties before (and yes, I fill out all 60-odd boxes so that labor & liberal go last in my preferences), having said that, I'm an extremely reluctant Greens voter. Most of their underlings seem incompetent, or are just as warped as any other career-politician, or they remind me of people I met at university who seem capable of instantly forming very loud, unshakable, annoying opinions on any subject regardless of their understanding, knowledge, and painfully obvious lack of research.
Then there's the state governments...
You really need to read Crockford's "Javascript: the good parts". You absolutely do make private methods and vars (ever noticed that you can't directly call jQuery's internal methods? Or TinyMCE's? Or any other major library/framework?)
He also makes the case that actually JS has more patterns to allow code re-use. That's why things like Mootools can even fake things that look like classical class inheritance patterns for you, if you really want to do that.
Check out http://www.crockford.com/javascript/inheritance.html and http://javascript.crockford.com/prototypal.html and http://www.slideshare.net/douglascrockford/javascript-the-good-parts-3292746
Most datasheets I've seen are 10V, and certainly the dis/charge current is always in mA...
Whereas trains use HIGH voltage and current AC. When building a capacitor, you have to fight against conflicting requirements: high density, high current, high voltage, stable against environmental changes (humidity/temperature), stable against aging characteristics, stable against voltage/charge (and voltage/charge rate), AC/RF response characteristics, dissipation ("leakage"), among other things. Double-layer capacitors ("ultra-caps") sacrifice maximum operating voltage and maximum discharge rate (current) for charge capacity/density.
You can use SQL for everything, or you can use the best tool for the job. Usually that's SQL, but the other choices aren't just for big data: graphdbs like neo4j allow you to efficiently query deeply, arbitrarily linked data in a way that just can't be done with SQL.
And then there's the triple/quad-stores like jena, 4store, virtuoso which allow you to answer questions from your data that you couldn't even begin to imagine doing with SQL.
So although I think these graph & "triple" DBs are the way of the future, it's an extremely crazy thing to ditch SQL for your core business data. But it's also crazy to dismiss these non-SQL technologies simply because they aren't SQL - there is a lot of potential in having your data in one or more of these "alternatives", similar to the value you get from wiring up solr/lucene for full-text search.
We've been doing a project with MongoDB (no, we're not using it as our authoritative data store, but neither are we using SQL for that), which we could have done with an SQL DB, but honestly we didn't have the resources to do so. We have less than a million records, hardly a huge amount of data - our data is arbitrarily structured, fitting the document-object model perfectly, and makes for absolutely meaningless SQL tables (with many joins, stored procedures so that we could try to service an obscure perlish query language).
Additionally, the ability for us to delegate some parts of the more complex queries to JS on the server(s) is incredibly useful. And the integration cost was very low: the perl driver is officially supported by 10gen, it fit quite naturally into the project... the proof of concept was done in just a couple of weeks.
Our contractor has, however, discovered some irony with MongoDB's "schemaless" claim: unless you can pick an extremely finite set of fields to be indexed, you can't actually do arbitrary queries on arbitrary documents... in other words you need to decide on a schema :-)
Thanks for the feedback. wl txpwr1 shows 28mW, as I set it from rc.common.
I know that the APs are broadcasting way too strongly by default with 7.09. At one site I was approached by one of the users asking why he could get a reasonable signal strength but not connect. He was trying to access it from a location that I know was previously unusable under White Russian. A few days later, even people close to the AP couldn't reliably stay connected.
By the time I returned to the site a week later, none of the APs are able to transmit at all. A site survey for instance, finds other APs, but no laptop, other AP, or my nifty wlan meter thing can see any trace of a signal whatsoever. Try reflashing back to Linksys firmware, no luck. Good as junk.
Had the same problem at the other site. These had more time to kill themselves though, as the network was unused for a few months (inter-office LAN connectivity project that was never utilised). I replaced cables and antennas until I discovered the APs themselves weren't transmitting (but were able to scan and report other networks via site survey). Rooftop/weatherproofed units too, not easy to replace.
It seems bizarre I'm the only one suffering this problem. I'm running 12V PoE over 20m of Cat-5e. Perhaps the DC-DC conversion after a 2 or 3v drop (~10V at the device) induces some kind of noise that stresses the transmitter circuitry somehow.
Who knows. I do know that I don't want my APs defaulting to 127mW transmit power.
It still would have been handy to check/confirm transmission power setting from what is otherwise a fantastic web interface on 7.09. It's simply impossible without installing the wl package, and this oversight surprises me.
My confidence in the platform was never all that high, but people are wary of spending the money on what I was using previously (Nexus Airpoints).
Please, do this test:
ipkg install wl
wl txpwr
I just set another WRT54GL up today, it reports 127mW. What does yours say?
I'm not sure what I'm doing different, but it always defaults to maximum without running wl at boot to set it otherwise.
The rest of Kamikaze is brilliant, which is why I persist with it. I've compiled my own firmware now so that wl and the modified init script is there by default (along with OpenVPN, knockd, QoS, etc).
I'm running WRT54GLs, v1.1, all from the same supplier. The unit I just configured has S/N: CL7B1H416073.
It takes about 3-7 days for the AP to finish killing itself. I installed four at one site, and two at another, then forgot about them as I usually do. User complaints were not immediate.
It's not the first time I've heard this, but it comes from people who've observed degraded performance after increasing their AP's power output (usually with a 3rd-party firmware).
What's going on here is that:
a) Clients are still transmitting at normal power, so the AP can't hear the clients.
b) Many APs are built with circuitry that doesn't like to be pushed very far beyond factory defaults with transmit power: the signal really does get "noisy" at high power settings.
Too bad the default Kamikaze 7.09 OpenWRT firmwares kills any and all (six!) WRT54GL routers that I put it on (previously ran White Russian brilliantly). Apparently flashing these things with said firmware out of the box defaults the output power to 150mW (default is 28mW), and fries the transmitter circuitry. There's no option to fix this, you're supposed to install a package onto the router called "wl" and hack a call to this utility in the init script for yourself that sets the output power at bootup.
Actually, wifi is more like an old-school hub (vs switch) than you think: all clients talk and listen on the same medium (radio vs wires) at the same time, the only difference being that wifi networks have the added problem of not being able to tell if they've caused a collision while transmitting.
To successfully share the bits around, wired hubs employ CSMA/CD, whereas wireless does CSMA/CA - Carrier Sense Multiple Access, Collision Detection vs Collision Avoidance.
As the number of communicating nodes increase, you get pretty much the same (drastic) drop in the total amount of data shifted (due to collisions/contention) as you did with wires (using a hub).
I've also mostly developed in C (on embedded systems), among other languages (Verilog on FPGAs, for example). At work, I chose Ruby (could have chosen Python, but found Ruby more expressive). I would have considered Java more seriously if the company I worked had more than one development engineer (me) and had a more rigorous engineering style (at the moment, virtually none). So I'm sort of hacking on the fly. It's a large app, for me, possibly the largest scope for anything I've undertaken before. But I could not have done it in C/C++/Java - customers don't know what they want, so much pie in the sky features they desire. With Ruby I've been able to meet deadlines and even implement their crazy ideas without as much pain as I would have had with Java. So there you go. Java can thrive in properly engineered, bureaucratic apps with the resources to support that deveopment style. Ruby thrives in a more ad-hoc, rapid development. You may even consider it a prototyping language (in fact, I have done that once: knocked up a prototype web interface for an embedded project for the customer to confirm what they wanted. Then it took two months to do the same thing in C). If you can't see the merits (and pitfalls) of where dynamic languages are useful, you need to play with them more ;-)
I've been tasked with coming up with a GUI that involves data visualisation and report presentation, where before I've mostly done very discrete back-end or embedded systems stuff.
Because there's real-time data visualisation (as well as historic stuff), I've heard about the Tufte books before and so bought all four available at bookware.com.au - Beautiful Evidence, Visual Display of Quantitative Information, Envisioning Information, and Visual Explanations.
Still waiting on them, probably won't be able to appreciate them all in time, but hopefully I can make my app suck less than the existing solutions I'm tasked to replace.
My application is loosely what might be traditionally known as SCADA... but for various reasons we're not using traditional SCADA packages. We're presenting industrial process data, traditionally there are real-time figures and "dials"/bar graph gauge type indicators, along with graph plots that resemble the paper and pen chart recorders this software replaced many years ago.
Any particular one of the four books that people might know to be most useful for me, or a suggested reading order anyone might have?
Because the statement that it would be equivalent to "The power output of the Sun for .39 nanoseconds" is misleading.
Don't get distracted by the 39ns figure. Power is an instantaneous quantity - it is a rate at which energy is transmitted. They are saying that the bomb sustained a level of power (rate of energy) output and held it there for a period of time - 39 ns - that approached 1% of the sun.
I repeat: 39ns is just the period of time that the power level peaked for. They calculated that the amplitude of the power peak itself, was equivalent to 1% the power output of the sun.
We don't care about how long the peak lasted for, the 39ns, unless you start integrating power over time as you just did, in which case you're comparing a quantity of energy, rather than a rate of energy output. Yes, I suppose you could say that 39ns @ 1% sun power is equivalent to an amount of energy produced by the sun in 0.39ns, but that's not the interesting number here, because we could similarly integrate just about any huge power source over a long enough interval of time (hours, days, years, whatever) to come up with "the same amount of energy output by the sun over 39 ns".
So the interesting number is in this case, yes, that the actual instantaneous absolute power output of the bomb approached 1% of that of the sun, albeit for only 39 ns.
Quite remarkable...
Oops. Posted that anon. Here I am.
Wow, you are being very picky today. Saying that "rising voltage does not blow fuses, rising current does" is irrelevant... There was nothing technically wrong with parent's observation on the behaviour of protection circuits when presented with excessive voltage. Who knows, a transient voltage suppressor may be protecting the load as well as that fuse, and the TVS may provide that current to make the fuse blow on over-voltage.
My background is admittedly low power engineering, albeit in hazardous areas (intrinsically safe systems where there's explosive gases in the air). As for utility power transmission lines, I'm pretty sure those fuses are usually designed so they don't spray shrapnel everywhere. Particularly at substations the line fuses are quite exotic, composite fuse links (involving silicon), immersed in spark-arresting gases and so on. On the LV side of some plants I've seen them use two different types of fuses in series - one to interrupt "dead-short" type over-current, and the other for more gentle 200% overload situations.
But all this talk of how fuses work is... redundant? These cabinets are using unstable batteries that do require precautions to maintain a safe working life. I can tell you that in IEC Zone 1 and Zone 2 hazardous areas (these might exist at mine sites, chemical processing plants, petroleum refineries, etc), it's very difficult to certify lithium batteries in this environment - they are just not trusted. A certified Lithium battery has electronics to protect each *cell* of the battery: monitoring internal pressure of the cell, temperature, preventing discharge below a safe voltage, preventing overcharge to above a safe voltage, and of course solid-state current limiting, as well as fused protection. And did I mention that's on each cell inside the battery? Because all it takes is one faulty cell to really, really ruin your day.
I don't understand what there is to discuss... unless someone blew it up with explosives, the unstable batteries in the compartment lacked sufficient internal protection circuitry to prevent explosion. Maybe a faulty charge circuit coupled with a dud thermal fuse in the battery (or all the batteries, if there was a faulty batch) caused all the cells to heat up, internal pressure exceeded casing maximum, it vented flammable material which ignited, starting a chain reaction by igniting the rest of the cells/batteries that were already just about boiling and ready to pop.
"Document Markup" is an interesting way of describing of
And yet, Microsoft prance around with the "Open" prefix. And yet, their RAND patent license excludes free software.
The difference being that Microsoft's spec has things like "do it the way Office 97 does it", and the ODF spec doesn't.
The simple fact that there are other Office suites already reading and writing ODF files other than OOo/StarOffice (Abiword, KOffice for example) demonstrates that it is a viable and workable standard.
It's my impression (others have read more of the 6,000 pages of documents than I have) that the same could not be successfully achieved from the OOXML spec.
That's funny, exactly what Microsoft seems to be planning. Their royalty free patent license may only be granted if you implement their standard EXACTLY (a herculean feat in itself). Want to enhance or modify your software, as the GPL explicitly sates you should be allowed to do? Sorry, you just agreed to get sued by Microsoft..
Actually, at my uni, having anything to do with the student union was a novelty that a very small minority engaged in first year, and after that... well, there isn't a single person in my engineering year (or below) who I know of that gives a damn. In fact, the student in my year who "won" the seat to represent the students within the engineering school was sort of forced upon them.
Now with VSU (voluntary student union), the student union is even more pathetic than it ever was.
Not that I really mind - my union fees (easily as much as 3 text books) should be spent on student services at the campuses, not on rudderless effigy burning protests that in the course of a few hours behind the megaphone apparently end up with 15 different agendas, introduced whimsically as if to keep themselves from getting bored, ranging from gay rights for albino homing pigeons to implying that our welfare system for full-time unemployed students is somehow below "the poverty line" (granted, austudy + rent assistance isn't decadence living, but I hate ignorant people using hyperbole like this because it trivialises the real issue they're using as an attention getter - and the reality is these idiots have never come close to trying to survive actual "poverty line" conditions as defined by the UN. For instance, I don't see any of the African or Indian students who have come over from very poor backgrounds on a life-changing scholarship programme, joining in their protest and agreeing with their cause).
Then there's the bus trips for protests out at woomera and embezzling e.g. print costs for the student magazine out to a mate's print shop instead of using much cheaper university print facilities... oh, and doing their damn job that some of the union positions were paid to do, instead of sitting around caring more about influencing "the system" (in any way possible) than the actual issues they use as a platform to accomplish this.
In principle, I know that unions, and student unions, are very useful when they're needed. But when they exist for the sake of existing, they invent their own causes to justify their existence, and somehow it breeds a culture of people who are in it for the "scene" more than anything else... Creating an attitude that alienates them from the vast majority of the student body who have more important things to worry about.
Good riddance to the SRC, RIP student services...
The issue, as far as I can gather, is that an attorney charged with protecting the electoral system against fraud, was actually working for an office which itself is challenging and invalidating millions of otherwise legal votes.
The E-mails gave specificity as to how this was done. I'm under the impression the sheer scale of vote invalidation should be setting off alarm bells, but maybe it's a normal number. Regardless, IIRC they have 70,000 names on the "caged" list. I suppose it could be an elaborate hoax to frame the Republicans - which seems to be what you are suggesting could be a possibility?
The company that received the E-mails erroneously is some anti-republican group, clearly - georgewbush.org is dripping with sarcasm.
There appear to be a lot of very high-profile people visibly upset at these E-mails. I haven't heard any suggestions that imply they could be fakes.
Either way - the sheer number of votes challenged, along with the rest of the circumstantial stuff (like, the president of the USA's entire E-mail system losing every E-mail the subpoena was interested in) makes me think the ball is more in the Republican's court to explain and defend their actions.
NB: I'm not from the USA. And I don't like the sounds of the democrats either, not that I'm an expert at having an opinion on US politics. Just sharing my view so far from what I can see.