What is, "there used to be a lot more matter and antimatter before they started canceling each other out and now we live amongst the debris"?
or, from my safety fifth-grader...
What is "the standard model is wrong"?
And I don't mean that in a bad way. The "flat earth" hypothesis was an _amazing_ deduction at its inception. It was only off by eight inches declination for every mile. This was a _tiny_ margin of error. But error compounds and so does any other form of tiny, so eight inches per mile, an error of ~.0126% (e.g. 8/63360) was enough to make the earth round.
Microsoft, and IBM and everybody else who is in any kind of position to be IP-trolled should _really_ be arguing that all software patents are invalid. So what if it thus invalidates their "investment" in software patents. They only spent that money to "repel boarders" in the IP wars anyway.
Funny thing that, the more you talk about patent trolling the more you get to analogies like "repel boarders". It's almost like "IP" stands for "Intellectual Piracy".
Hey big companies! Stop skirmishing and fight the war. Outlaw the Intellectual Piracy that matters. Outlaw the USPTO (at least in your bailiwick.)
They only elected GWB once. The first time the Supreme Court elected him by saying speed was more important than accuracy in voting.
As a side note, the first election of GWB was the first time in history that a Republican got 100% of the black vote in the venue of record...
And Again, with less invective: the first election we didn't really know what a disaster GWB would have been as a president. If he haddent got a bump because of his own incompetence basically allowing 9/11 he would have been on vacation the whole time. It's the second election that I find so fascinating. It was the electoral equivalent of throwing good money after bad. Kinda the electoral precursor mindset that the sub prime events pivoted upon.
Then again, this is the second cycle of the same mess, so I register no surprise at all.
Here are the things you probably didn't look into...
1) If you control the app source code, adding fadvise() calls after open to tell the application that you are going to use random access will turn off the read-ahead on those files selectively. The real reason that you are seeing the random access kill you is because you are probably using the default read ahead of 128k for the partition. That means that if you are reading 1k records, for example, and the total size of your active file set is greater than your dchace memory, your effective raw read length is 129k and you are wastind nearly 100% of that read.
2) For the reasons set forth above, go into the various/sys/block/(whatever)/queue/ directories and set the kb_read_ahead values to zero, or maybe 8k for the partitions where you are storing these large random access files. Remember that in your case there are layers (the raw block device, and the raid, etc) and tuning the layers may be in order.
3) The optimal stacking order is LVM on top of dmcrypt on top of raid_5_or_6 on top of media. No I don't say to do all of that every time, but that's the optimal order. If, for instance you put the raid over the encrypted volume then you will pay about 3 times the encryption cost than having the dmcrypt over the raid. (this is because saving and computing the xors of data that is already encrypted is much cheaper than decrypting all the sectors, computing the xors, then separately encrypting all those sectors as they return to the media.
4) for large storage solutions _always_ have an LVM on top. The overhead cost of a volume in a volume group is basically a single add and an extra function call. In exchange you can persistently set the readahead profile of each partition. You can also migrate your elements between physical devices safely and on the fly. All the reasons why aren't terribly obvious until the day it saves your life, or your weekend. (I have, for instance, plugged six USB hard drives into a system with a failing raid, built a raid on those USB drives, added the new raid to the existing volume group, and then migrated the active partition from the failing raid to the plugable raid with the system up the whole time then Dropped the failing raid from the volume group. Meanwhile I built a second computer to replace the first. Shut down the old computer. Moved the plugables over to the new one. Booted up into production. then built the new permanent home for the partition. Added that raid into the volume group. Migrated the partition, then dropped the plugables. Total down time was the cut-over between the two computers. I also then got to grow the file system onto the new raid in the new computer (the drives were bigger) during the next maintenance window. LVM is very much your friend.
5) it is easy and even desirable to build a volume group that is over heterogeneous storage. In particular lets say I have a raid6 built over five drives/dev/sd[a-e]1 called/dev/mapper/main. I will make a volume group "system" over/dev/mapper/main and/dev/sdf1 and then build my runtime file system in an LVM that is only on the/dev/mapper/main part of system. Now, when I have to make backups etc I can create the snapshot in the/dev/sdf1 part of system without seriously impacting my operations by competing for raid computation time/space.
6) make sure you tune the stripe cache for any raid to a big number if you are doing writes, particularly random writes, to it, the default stripe cache is tiny. This is also important for keeping your read rates up if you end up in a degraded state.
7) take a serious look at your choice of scheduler.
So anyway, the degradation you describe is identical to having an improperly (or naively) tuned storage stack. In particular it sounds like read-ahead waste.
Here are the things you probably didn't look into...
1) If you control the app source code, adding fadvise() calls after open to tell the application that you are going to use random access will turn off the read-ahead on those files selectively. The real reason that you are seeing the random access kill you is because you are probably using the defalut read ahead of 128k for the parition. That means that if you are reading 1k records, for example, and the total size of your active file set is greater than your dchace memory, your effective raw read lenght is 129k and you are wastind nearly 100% of that read.
2) For the reasons set forth above, go into the various/sys/block/(whatever)/queue/ directories and set the kb_read_ahead values to zero, or maybe 8k for the partitions where you are storing these large random access files. Remember that in your case there are layers (the raw block device, and the raid, etc) and tuning the layers may be in order.
3) The optimal stacking order is LVM on top of dmcrypt on top of raid_5_or_6 on top of media. No I don't say to do all of that every time, but that's the optimal order. If, for instance you put the raid over the encrypted volume then you will pay about 3 times the encryption cost than having the dmcrypt over the raid. (this is because saving and computing the xors of data that is already encrypted is much cheaper than decrypting all the sectors, computing the xors, then separately encrypting all those sectors as they return to the media.
4) for large storage soulutions _always_ have an LVM on top. The overhead cost of a volume in a volume group is basically a single add and an extra function call. In exchange you can persistently set the readahead profile of each partition. You can also migrate your elements between physical devices safely and on the fly. All the reasons why aren't terribly obvious until the day it saves your life, or your weekend. (I have, for instance, plugged six USB hard drives into a system with a failing raid, built a raid on those USB drives, added the new raid to the existing volume group, and then migrated the active partition from the failing raid to the plugable raid with the system up the whole time then Dropped the failing raid from the volume group. Meanwhile I built a second computer to replace the first. Shut down the old computer. Moved the plugables over to the new one. Booted up into production. then built the new perminant home for the partition. Added that raid into the volume group. Migrated the partiton, then dropped the plugables. Total down time was the cutover between the two computers. I also then got to grow the file system onto the new raid in the new computer (the drives were bigger) during the next maintence window. LVM is very much your friend.
5) it is easy and even desireable to build a volume group that is over heterogeneous storage. In particular lets say I have a raid6 built over five drives/dev/sd[a-e]1 called/dev/mapper/main. I will make a volume group "system" over/dev/mapper/main and/dev/sdf1 and then build my runtime file system in an LVM that is only on the/dev/mapper/main part of system. Now, when I have to make backups etc I can create the snapshot in the/dev/sdf1 part of system without seriously impacting my operations by compeeting for raid computation time/space.
6) make sure you tune the stripe cache for any raid to a big number if you are doing writes, particularly random writes, to it, the default stripe cache is tiny. This is also important for keeping your read rates up if you end up in a degraded state.
7) take a serious look at your choice of scheduler.
So anyway, the degradation you describe is identical to having an improperly (or naively) tuned storage stack. In particular it sounds like read-ahead waste.
My my, admittedly lay, understanding the prohibition of making images of Muhammad was to prevent his elevation to a protected status. That is, it is to prevent Muhamud from becoming the target of Idolatry.
By treating his image as special and sacrosanct, Comedy Central has violated and insulted the tennants of Islam by putting the prophet before Alah.
This homage to his image is _literally_ a damned if you don't position.
Sure, a lot of "Arab Rednecks" who don't understand their own religion think an image of Muhammad is wrong. But given that making _fun_ of Muhammad is in no danger of raising his image as an idol.
So technically Comedy Central has attacked Islam with their censorship.
(but since neither position makes any sense at all, perhaps they should stop pretending that bowing to a vocla minority of any persuasion is just wrong.)
In practical terms I would think they would be more afraid putting the dick to L Ron, since those Xenu Fearing Wing Nuts are way more likely to make their life difficult.
No so much vampires. but you have a point about how it might turn out. I'd think of the prion as the ultimate sexually transmitted (in the medical sense here, not the common usage) disease. But hey, I am trying to leave as much of the original story structure intact with these suggestions.
My favorite is the truely profoundly mind-breakingly alien thing. Note that I didn't advocate adding aliens themselves, just the artifact. Like a space station or a communication device that drives humans mad.
A real life friend asked about the reference so...
If you watch the alternate endings on the DVD of 28 days later, they wanted to have a symmetrical ending where the main male lead would sacrifice himself to save the girl's father. Medically they wanted to "cure" to be a huge sacrifice, where it takes one person surrendering their life to save any other from the disease.
The ending scene would then have the male lead tied down to the table watching the scenes of violence on the monitors, in a poetic juxtaposition to the monkey in the opening scene.
They realized they "couldn't do that" because the disease was too infectious for them to do the whol-body blood transfusion they envisioned. They were, of course correct, no transfusion could do it...
So here is the "patch" for the movie to give them their ending, both visually and "technically" at least to the degree of "technically correct" established by the rest of the movie.
PATCH: The cure is basically the original virus that was engineered to become the super weapon. You give that to an uninfected donor person. Then as they go mad the original, slow way you give them a shot of bone marrow collected from the person you want to cure. This contains the infected immune cells of the intended recipient. The donor's immune system, primed with the more robust but slower original virus produces antibodies to unique viral strain created by the fast virus and the recipients genetics. The donor's blood serum now contains an antibody uniquely matched to the recipient which will jump-start their immune system. The donor cannot be re-used since introducing a second bone marrow sample would produce an immune complex that would simply kill the second recipient systemic organ rejection.
Okay, its all techno-babble, but it's _good_ techno-babble. It can be acted out on screen with big pantomime gestures. I has all the dramatic elements the writers and cinematographers were after. It actually makes more sense than a disease so virulent that it can drive you mad in seconds.
But that is the essence of a good patch, it changes the minimum while correcting the flaw and allowing for the desired outcome.
Apparently you have blotted out part of the movie to save yourself from the cognitive dissonance the horrible hidden planet engendered within your soul.
The planet was hidden. Go back and look at the scene when they are looking for the name Miranda and then looking at the empty orbital plot. How they puzzle about how there is nothing there. etc.
Since, by definition, it wasn't "in Reaver territory" before there were Reavers, there were dozens if not hundreds of years for people to go there when Miranda was just the outer-most planet in the system. Go there often. Perhaps weeky, delivering construction supplies. Inhabit it at least as well as they can inhabit empty space. Use it as a military base or a smugglers hide out. Use its gravity in slingshot maneuvers to get between the second and third outward most planets as they fulfilled the 500-year orbits dictated by the planets being far enough away from a main-sequence star so big it can afford a good dozens planets a stable orbit in its "habitable zone".
Or otherwise we hide this planet while the adventurous pioneering souls of spacefaring civilization all collectively forget the invention of the telescope and generally stop looking at the lights that move in the night sky.
And if you have ever hung out with people who go out to sea and make it back, were the place inherently bad prior to the Pax and the Reavers, then there would be drunken bar stories and warning beacons and all manner of record on the existence of the planet and its exact condition.
Not to even begin on the topic of all those dead people having families, heirs, and bank accounts; and some of them being nascient brown-coats or who went to college with investigative journalists; or who owed big money to that evil Russian Dude...
Don laugh at Y2k. It was serious. Not the media part, but the actual washing of all that COBOL and PL/1 and all those terrible boundary cases. Much of the Y2k money was well spent.
But paying good salaries for no benefit, and DRM is _no_ benefit, sucks our money into a vast entropy sink. Sure some marketing execs, corporate execs, and charlatans skim a little off the flow of your cash into nothing. Notice how all those three profiteers have one thing in common? (okay, maybe several, but none of them are "smart producers of lasting value").
DRM is theft. An economic drag. A criminal dissipation of resources in pursuit of the idea the _totally_ _bizarre_ notion that there are a select few who deserve a lifetime of income for a days worth of work.
There, I said it. I think that copyright (the right to keep someone from bastardizing your work) should continue for the lifetime of an artist, but the charge-for-it-right ought to basically evaporate once everyone has made a decent wage for the time invested plus maybe 10% profit.
In what sane universe would I be paying Peter Frampton (deceased) or his heirs (all also deceased) for a performance of "Do you feel like I feel" if the purpose of copyright is, as stated, to encourage Peter to make more music (aparently from the beyond), while that "encouragement" happens by paying a record executive's lawyer to sue my dead grandmother?
Its all crazy talk. (and I am an author, so I understand the desire to keep the furries out of my mythos, but I don't see that the grand children of my nieces and nephews deserve to "live off of" me writing one book. [see Gone with the Wind.])
Coulda, sure, but the script and mythology already have habitable planets and moons, and even with a very, very, very wide/deep habitable zone, there is a minimum distance apart that 1-gravity-or-so planets would have to be separated by in order for them _neither_ become one large N-gravity planet _or_ pull each other into bits. After that they would have to be even further apart to keep them from giving each other gravity boosts into some very unlivable orbits.
So no, there is no excuse. Since their habitable planets have some habitable moons, you now have moons with moons of their own, all at about 1-gravity, orbiting a very huge and very distant sun so that you have a habitable zone as wide as the zone from earth to about neptune and about as far away from the central gravity well such that the nearest planet is about as far from the primary as (pulling a distance from my ass without any math at all 8-) Pluto is from Sol. (I make that totally wild guess just by trying straight scale-up the width of our habitable zone to carry seven plus planets of about earth normal gravity, Remembering that Venus is earth gravity or so but Mars is not, and both are marginally outside the prime habitable zone.)
That said, Making the Pax transmissible is super easy since I am editing the whoel ideal wholescale. It wasn't a "gas" it was a particulate aerosol introduced in the air handling system of a central facility (not the terraforming facilty plant) (or just sprayed from an airborne craft). In the field, possibly because of the idiosyncrasies we know every teraformed world developed (see "The Train Job") caused a prion disease (improperly folded protein). In most people the prion caused an amplified pacifying effect by destroying the aggression centers of the brain the original drug targeted. In some it destroyed the inhibition centers instead. The particulate is long gone, but the prions persist. Close contact with a reaver can infect you with the prion, especially if they maybe do the raping and maybe some of the eating, but then don't get around to the killing.
It's all fixable. Almost.
You just do _not_ get to "hide" a planet in the last minutes of the the last 1/5th or less of 100 or more years of planetary occupancy. If nothing else smugglers and criminals will occupy any gravity well that the Alliance doesn't. So a failed terraform would lead to a nice juicy "safe but abandoned" place, with a planet to hide behind and mine and whatnot, as resources and safety for, say, a brown-coat base during a civil war....?
Miranda, as presented in the movie, was just lame and was a total speed bump for my suspension of disbelief. I got over it to enjoy the rest of the movie.
Between the "one solar system" and "the secret planet" the movie took the mythos from "workable" to "cheap framework for space opera". Still loved the characters and the presentation, but they could have just called the whole place 'the forbidden zone' after pulling that much completely unnecessary mcguffen out of its ass.
I agree that Miranda would have been fine in a system where they weren't all orbiting one star.
And I would give you Reavers being the result of Pax if it were said that, perhaps, Pax was or caused a prion disease of the brain like Mad Cow disease etc. Easily done there.
Then again, the less-than-twenty (and likely less than ten) year timescale would belie Reavers being an old wives tale.
Draw out the time scale on paper sometime. Reavers would have been absolute fact, and a new fact at that, if the movie is cannon. They'd have exploded across the whole star system pervasively, and been hunted to extinction. They just wouldn't have the longevity (what with unshielded reactors and having to replenish their ranks and food supplies, since there isn't a 'reaver farmer' growing grain and we know food is scarce in that reality etc. In the absence of a reaver economy the would have been starved out of orbit at Miranda in like a month. etc.)
The whole thing reeks of vampire fan-boy logic. Hundereds of vampires inhabit 1840 Paris/Moscow/New York/New Orleans (as they have for hundreds of years etc), each one draining the blood from one human each night, year in and year out. So even in WWII where people die and go missing all the time 365.24 * 100 is, on average 36,524 people gone missing from one city each year. The numbers just don't work if you even glance sideways at the math.
To fix the problem with least effort: Keep the Pax, but try it on a cruise ship or a space station (e.g. "The Miranda" instead of "The Planet Miranda", or hell on a couple dozen people in a lab. Start with a smaller population. Make it properly contagious like the aforementioned prion disease (raver decides to rape you first, or feed you some of his flesh or blood, and _maybe_ you convert if you live long enough or get left behind) instead of a mystically repeatable experience that somehow has a predictable morphology, common enough for Mal to know the exact progression well enough to narrate it 8-). Same evil government, same evil plot, harder to come up with an armada of Ravers granted, but workable. Hell you even get to cut a word and some "look at the nothing" computer interface mock-up CGI from the script. Everybody wins. 8-)
Terraformign didn't take? Given that the war was there, and before the war there was a how many years or centuries for the aliance to grow up, grow dispised, grow dispotic, and engender warfare; well the system has been around a mighty long time. Everybody would have knows Miranda (the planet) was habitable long before the need to develop the Pax came about.
Hell, the firefly class being "good for smuggling" means that there was smuggling, which means that if Miranda (the planet) was ideally terraformed then everybody would have been there, and if it were a teraforming disaster, all the people who didn't want to be where all the people were woudl use it as a base just like they do everything else.
In the scale of things, having Miranda go unnoticed and discovered the way it was has the credibility of Atlantis. That is, it is as likely as finding out on the news this eventing that a land mass the size of Australia was just "discovered" _anywhere_ on the earth, and it had been hidden by the UN because 20 years ago they decided they needed a secret continent of their very own.
You just cannot hide a planet. And the _planet_ was hidden (watch the movie again, look at all that "nothing" in the CGI rendering of the not-the-internet when the discover the planet). yes, planet.
There just is no excusing that whole planet thing. Hell I came up with a way for the 28 Days Later people to pull off their originally preferred ending (watch the DVD extras) and that was _easy_ in that mythos. But outside of Buck Rodgers, circa 1930, you just _don't_ get license to hide a whole freaking planet co-orbiting a star occupied by multi-planetary spacefaring society full of smugglers and rebels.
Now, to be honest, when required to argue the other side in a different context, I took the position that the perturbations of navigation were hidden in the navigation system software. Nobody would have any cause to double-check the computer as long as the easy-to-computer compensations worked and everybody ended up at their destinations.
But I was defeated by the ten year old with binoculars. There is no way that a planetary mass, habitable and life sustaining at that, let alone one _surrounded_ by a cloud of reaver ships has an albedo low enough to hide, or even go unnoticed traversing the night sky of a few dozen planets and moons.
I love Joss' imagination. His vision is good. Like any bond villian, he needed to find him 5 year old to bounce his evil plan off of before putting it to paper. There is no excuse for Miranda.
1) from the series, yep, some people go crazy when they spend too long on the edge of that much nothing.
2) in the series it was mentioned that, despite the rumors, there have never actually been any aliens or alien artifacts found. What if there were something so alien way, way out there, but attractive enough to draw people in, that was so alien that it broke the mind. (Would have dove-tailed in with River being prescient and mad, suppose the blue-hands were working from "the only reliable evidence" ever beamed back, trying to unlock the secrets and abilities of the thing without bringing on reaver madness.)
3) Evil government experimentation, version "not-dumb", something infectious. Call it Pax if you want. Give it the same reason and history, but instead of a "chemical" or in _addition_ to it being a chemical have raverdom be an Infectious Prion form of the original chemical (see mad cow disease). Still need to drop Miranda and the one central sun, but now being force-fed a little reaver flesh would maybe make you one of the family. Hell, "The Miranda" could have been a freaking cruise ship and the thing would work, but not a planet.
At a minimum, some serious retcon needs to take place to repair the plot damage the movie inflicted on the franchise.
I just wish they would ask me about this stuff before the script goes into production. 8-)
Oh, and get rid of the whole Miranda bullshit. The people who ply the lanes of space would neither "overlook" nor "forget" an entire main planet over the course of less than 20 years. Nor could such a thing be hidden as, outer-most or not, it would show up on everybody's orbital computations as a huge perturbation in their plots. Let alone one ten-year-old with binoculars.
Oh yea, and drop that whole "all the planets orbiting one sun" nonsense since it isn't workable. Miranda would have been frozen ice-ball _or_ the "inner planets" would be molten slag.
Don't get me wrong, I loved the show. The movie needs to be declared out-of-cannon before the series would be workable.
I could have come up with a better "reason for the reavers" in my sleep. The original one from the series (mental erosion from facing the emptiness of space etc) was good enough. Hell, the movie contradicted the series directly. If the Pax caused reaverdom, the the episode where the one guy got tortured and became a reaver himself woudln't have worked unless the reavers carry a supply of the otherwise secret Pax around and deliberately pre-expose potential recruits to it before deciding who to kill, rape, and eat (in that order, if you're really lucky).
DRM is the software version of Perpetual Motion. It is simply not possible to make the device described work as intended. But because of "the enormous potential income" should someone succeed, the greedy interest keep flushing money into the pockets of charlatans and charging the populace a tax for their stupid avarice.
Since DRM can only work if all the parts of the system are controlled by external DRM, including all the DRM enforcement parts, you end up with "its elephants all the way down."
So we will never be done until it is simply illegal. Just like the patent office will not accept patent applications for perpetual motion machines, and the FDA will not let unproved drugs out into the wild (in theory anyway 8-), the FTC (etc) will eventually need to refuse to let people try to sell things with this snake oil in it.
But like those remedies and limits, it will take a couple hundred years of corpses and bankruptcies cause by the offensive practice of duping companies into "DRM" before anybody finally acts to stop the scam.
And even then, people will still try to sneak it in the back door as "holistic systems engineering" or whatever.
You _aren't_ going to get a key for "The White Album" you are going to get a key for "the 2011 release of 'The White Album" in MP3 format from Sony Interactive for use on sPlayer #xxxxxxx" simply because they _can_ be that specific and they _don't_ want to sell anything once that they can sell a million times.
DRM == RENT, and illegal prior restraint, and a scheme that can never actually work because it is a system that violates every principle of both software engineering and cryptography. No matter how you slice it, DRM is a stupid waste of leptons, time, and money. It is a system based on a complete lack of modularity and locality.
DRM is a classic case of "who will watch the watchers?" and not just at the corporate and financial and cultural levels. As a simple exercise in software engineering DRM must fail. It is a system that must be part of every element of a system (which is the failure of locality and modularity etc) to the degree that you need to have DRM policing the DRM system.
DRM is the Perpetual Motion of Software. People keep inventing new versions of it that don't quite work because no version of it can _ever_ deliver what is promised. Companies keep buying into the hype because they are blinded by "the potential". The only difference is that we are all being forced to buy these perpetual motion machines. Sure _this_ one has a battery in it, _that_ one has to be hooked up to the electrical mains. Some other one needs a waterwheel or a solar panel, and they will all tear off an arm or crush your child if you aren't careful... but we are _almost_ there... just one more scheme and we'll have it right...
The whole thing is a tax, levied by the stupid, paid by the sheep, and ready to break businesses when, I don't know, say Microsoft (or whomever) forgets to update a certificate (or whatever) before it expires (or whatever).
This is the kind of error messages you see from programmers who use exceptions _poorly_. I use exceptions well, and in doing so I don't just use the pointlessly empty exception classes. I have a tidy little toolbox with a slelect few exception classes that are normally instanced in a way that includes file and line numbers (for me) and useful text (for the user) at a minimum. Heck, when I compile with a debug macro defined (q.v. -DDEBUG) on a GCC/GlibC based system, my exception constructor saves a whole stack trace created at construction using backtrace().
Oddly enough, such exceptions are typically thrown in response to "doing error checking within the routine".
They are also damn handy inside of libraries where, for instance, you want to have a function that returns int and there is no "invalid" value to return (such as -1) because the whole domain of int is a valid value. (as in doing checksums or hash tables and so on).
The fact of the matter is, many people should not be trusted with power tools, and many people should not be trusted with manual tools, but nobody who thinks either kind of tool should not exist for whatever reason should be trusted with any tools at all.
You hate exceptions because either (a) you don't know how to use them properly or (b) you have been forced to use the code of someone who doesn't know how to use them properly, and perhaps (c) both of the above.
The real predicate is whether the condition can be properly handled locally or not. Passing an error return code (q.v. minus-one etc) back through ten layers of function calls is just as information poor a thing to do as throw an empty class as an exception. In fact its _worse_ in most cases as you need to make a tangle of logic to turn a callee's -1+errno for out-of-bounds into your local -1+errno for could-not-allocate, into the callers -1 pool-is-full-use-a-malloc-instead. This just multiplies for the full Cartesian Product of all possible error paths including those paths that also then fail.
Tell me true (presuming you use C/C++ on a posix-like os)...
Do you always check the return value from snprintf() to make sure that you didn't run out of buffer during conversion? Let me guess, you just choose to use really big buffers...
Do you wrap every write() in a pointer-increment loop to naturally catch when a write of ten bytes actually only writes 5. Let me guess, you only check for -1, and only on file descriptors that you think are "likely" to error out...
IMHO people who discard features whole-scale, and cite "hate" as a reason are generally guilty of practicing outside their ability.
There are plenty of things I "don't do", like I never use pure virtual functions in C++. I feel they serve no useful purpose except to annoy the programmer that comes along after me. (Having had a thread safety problem in a code block where one thread called a virtual function on a object that was being destroyed by another thread, and having the resultant call become a call to that NULL kinda irked me. But I learned my lesson. I put asserts or exceptions in the functions and document them, and document the class as having no useful default implementation. Also I've seen fat interfaces where the first thing a user has to do is stub-out the interface in order to make a single test call; where that lead to an important function ending up containing just "return 0;" all the way to production; which cost us time and money. But I don't "hate" pure virtual functions for finding them pointless and unhelpful.)
Er... Java is optional, and only required if you use the database engine (nobody does, because almost nobody knows when _not_ to use a spreadsheet, IMHO of course 8-) or some of accessibility and wizard thingies.
One of the ACs gives the actual quote and reference.
Plus OpenOffice.org (and I think core open office as well) dumped the larger desktop interface a long, long time ago.
Try something recent, and try reading the documentation, before you rail against any product.
Microsoft gets people to update by giving their product to the CEOs and "bigwigs". When everybody _else_ in the organization cannot read or use the new format for the documents, they have to keep bouncing transfered documents back to the aforementioned bigwigs. Eventually the bigwigs get tired of the fact that they cannot understand how to use save-as-older-format, and they dislike having their underlings telling them to do things, and they cannot bear to find all the files they saved and re-save them before they downgrade back to the old version... So the entire company naturally has to pay to upgrade everyone.
Repeat that at the border of the company. Every iteration of Little Company that works with and is dependent on Big Company, cannot allow themselves to be seen as unhelpful nor out of date, and they cannot bounce the documents they receive via email etc. without giving that exact impression...
Letting certificates expire is _not_ a Microsoft "strategy", it's an artifact of their adoption of "We don't care. We don't have to. We're The Phone Company" where there is no longer just one phone company, but Microsoft wants to be "The Software Company".
This _is_ egg on their face, but the only ones who will not yell "brilliant omelet" are the people who can connect the "Trusted Computing" dots. Letting the world _again_ see what it means to leave the keys to your property in the hands of any entity that doesn't _have_ to care is just another Microwhoops...
Slashdot Mirror
What is, "there used to be a lot more matter and antimatter before they started canceling each other out and now we live amongst the debris"?
or, from my safety fifth-grader...
What is "the standard model is wrong"?
And I don't mean that in a bad way. The "flat earth" hypothesis was an _amazing_ deduction at its inception. It was only off by eight inches declination for every mile. This was a _tiny_ margin of error. But error compounds and so does any other form of tiny, so eight inches per mile, an error of ~.0126% (e.g. 8/63360) was enough to make the earth round.
Ta dah! 8-)
Microsoft, and IBM and everybody else who is in any kind of position to be IP-trolled should _really_ be arguing that all software patents are invalid. So what if it thus invalidates their "investment" in software patents. They only spent that money to "repel boarders" in the IP wars anyway.
Funny thing that, the more you talk about patent trolling the more you get to analogies like "repel boarders". It's almost like "IP" stands for "Intellectual Piracy".
Hey big companies! Stop skirmishing and fight the war. Outlaw the Intellectual Piracy that matters. Outlaw the USPTO (at least in your bailiwick.)
Fixing that for you(TM)...
When the government is made into an arm of the RIAA and MPAA.
Okay, I guess it works both ways...
They only elected GWB once. The first time the Supreme Court elected him by saying speed was more important than accuracy in voting.
As a side note, the first election of GWB was the first time in history that a Republican got 100% of the black vote in the venue of record...
And Again, with less invective: the first election we didn't really know what a disaster GWB would have been as a president. If he haddent got a bump because of his own incompetence basically allowing 9/11 he would have been on vacation the whole time. It's the second election that I find so fascinating. It was the electoral equivalent of throwing good money after bad. Kinda the electoral precursor mindset that the sub prime events pivoted upon.
Then again, this is the second cycle of the same mess, so I register no surprise at all.
sorry, it gave me an error the first time... the second copy is even spell checked... 8-)
Here are the things you probably didn't look into...
1) If you control the app source code, adding fadvise() calls after open to tell the application that you are going to use random access will turn off the read-ahead on those files selectively. The real reason that you are seeing the random access kill you is because you are probably using the default read ahead of 128k for the partition. That means that if you are reading 1k records, for example, and the total size of your active file set is greater than your dchace memory, your effective raw read length is 129k and you are wastind nearly 100% of that read.
2) For the reasons set forth above, go into the various /sys/block/(whatever)/queue/ directories and set the kb_read_ahead values to zero, or maybe 8k for the partitions where you are storing these large random access files. Remember that in your case there are layers (the raw block device, and the raid, etc) and tuning the layers may be in order.
3) The optimal stacking order is LVM on top of dmcrypt on top of raid_5_or_6 on top of media. No I don't say to do all of that every time, but that's the optimal order. If, for instance you put the raid over the encrypted volume then you will pay about 3 times the encryption cost than having the dmcrypt over the raid. (this is because saving and computing the xors of data that is already encrypted is much cheaper than decrypting all the sectors, computing the xors, then separately encrypting all those sectors as they return to the media.
4) for large storage solutions _always_ have an LVM on top. The overhead cost of a volume in a volume group is basically a single add and an extra function call. In exchange you can persistently set the readahead profile of each partition. You can also migrate your elements between physical devices safely and on the fly. All the reasons why aren't terribly obvious until the day it saves your life, or your weekend. (I have, for instance, plugged six USB hard drives into a system with a failing raid, built a raid on those USB drives, added the new raid to the existing volume group, and then migrated the active partition from the failing raid to the plugable raid with the system up the whole time then Dropped the failing raid from the volume group. Meanwhile I built a second computer to replace the first. Shut down the old computer. Moved the plugables over to the new one. Booted up into production. then built the new permanent home for the partition. Added that raid into the volume group. Migrated the partition, then dropped the plugables. Total down time was the cut-over between the two computers. I also then got to grow the file system onto the new raid in the new computer (the drives were bigger) during the next maintenance window. LVM is very much your friend.
5) it is easy and even desirable to build a volume group that is over heterogeneous storage. In particular lets say I have a raid6 built over five drives /dev/sd[a-e]1 called /dev/mapper/main. I will make a volume group "system" over /dev/mapper/main and /dev/sdf1 and then build my runtime file system in an LVM that is only on the /dev/mapper/main part of system. Now, when I have to make backups etc I can create the snapshot in the /dev/sdf1 part of system without seriously impacting my operations by competing for raid computation time/space.
6) make sure you tune the stripe cache for any raid to a big number if you are doing writes, particularly random writes, to it, the default stripe cache is tiny. This is also important for keeping your read rates up if you end up in a degraded state.
7) take a serious look at your choice of scheduler.
So anyway, the degradation you describe is identical to having an improperly (or naively) tuned storage stack. In particular it sounds like read-ahead waste.
Here are the things you probably didn't look into...
1) If you control the app source code, adding fadvise() calls after open to tell the application that you are going to use random access will turn off the read-ahead on those files selectively. The real reason that you are seeing the random access kill you is because you are probably using the defalut read ahead of 128k for the parition. That means that if you are reading 1k records, for example, and the total size of your active file set is greater than your dchace memory, your effective raw read lenght is 129k and you are wastind nearly 100% of that read.
2) For the reasons set forth above, go into the various /sys/block/(whatever)/queue/ directories and set the kb_read_ahead values to zero, or maybe 8k for the partitions where you are storing these large random access files. Remember that in your case there are layers (the raw block device, and the raid, etc) and tuning the layers may be in order.
3) The optimal stacking order is LVM on top of dmcrypt on top of raid_5_or_6 on top of media. No I don't say to do all of that every time, but that's the optimal order. If, for instance you put the raid over the encrypted volume then you will pay about 3 times the encryption cost than having the dmcrypt over the raid. (this is because saving and computing the xors of data that is already encrypted is much cheaper than decrypting all the sectors, computing the xors, then separately encrypting all those sectors as they return to the media.
4) for large storage soulutions _always_ have an LVM on top. The overhead cost of a volume in a volume group is basically a single add and an extra function call. In exchange you can persistently set the readahead profile of each partition. You can also migrate your elements between physical devices safely and on the fly. All the reasons why aren't terribly obvious until the day it saves your life, or your weekend. (I have, for instance, plugged six USB hard drives into a system with a failing raid, built a raid on those USB drives, added the new raid to the existing volume group, and then migrated the active partition from the failing raid to the plugable raid with the system up the whole time then Dropped the failing raid from the volume group. Meanwhile I built a second computer to replace the first. Shut down the old computer. Moved the plugables over to the new one. Booted up into production. then built the new perminant home for the partition. Added that raid into the volume group. Migrated the partiton, then dropped the plugables. Total down time was the cutover between the two computers. I also then got to grow the file system onto the new raid in the new computer (the drives were bigger) during the next maintence window. LVM is very much your friend.
5) it is easy and even desireable to build a volume group that is over heterogeneous storage. In particular lets say I have a raid6 built over five drives /dev/sd[a-e]1 called /dev/mapper/main. I will make a volume group "system" over /dev/mapper/main and /dev/sdf1 and then build my runtime file system in an LVM that is only on the /dev/mapper/main part of system. Now, when I have to make backups etc I can create the snapshot in the /dev/sdf1 part of system without seriously impacting my operations by compeeting for raid computation time/space.
6) make sure you tune the stripe cache for any raid to a big number if you are doing writes, particularly random writes, to it, the default stripe cache is tiny. This is also important for keeping your read rates up if you end up in a degraded state.
7) take a serious look at your choice of scheduler.
So anyway, the degradation you describe is identical to having an improperly (or naively) tuned storage stack. In particular it sounds like read-ahead waste.
My my, admittedly lay, understanding the prohibition of making images of Muhammad was to prevent his elevation to a protected status. That is, it is to prevent Muhamud from becoming the target of Idolatry.
By treating his image as special and sacrosanct, Comedy Central has violated and insulted the tennants of Islam by putting the prophet before Alah.
This homage to his image is _literally_ a damned if you don't position.
Sure, a lot of "Arab Rednecks" who don't understand their own religion think an image of Muhammad is wrong. But given that making _fun_ of Muhammad is in no danger of raising his image as an idol.
So technically Comedy Central has attacked Islam with their censorship.
(but since neither position makes any sense at all, perhaps they should stop pretending that bowing to a vocla minority of any persuasion is just wrong.)
In practical terms I would think they would be more afraid putting the dick to L Ron, since those Xenu Fearing Wing Nuts are way more likely to make their life difficult.
oh... I see what you did there...
You make the U.S. Government sound like Microsoft... That's just mean...
No so much vampires. but you have a point about how it might turn out. I'd think of the prion as the ultimate sexually transmitted (in the medical sense here, not the common usage) disease. But hey, I am trying to leave as much of the original story structure intact with these suggestions.
My favorite is the truely profoundly mind-breakingly alien thing. Note that I didn't advocate adding aliens themselves, just the artifact. Like a space station or a communication device that drives humans mad.
A real life friend asked about the reference so...
If you watch the alternate endings on the DVD of 28 days later, they wanted to have a symmetrical ending where the main male lead would sacrifice himself to save the girl's father. Medically they wanted to "cure" to be a huge sacrifice, where it takes one person surrendering their life to save any other from the disease.
The ending scene would then have the male lead tied down to the table watching the scenes of violence on the monitors, in a poetic juxtaposition to the monkey in the opening scene.
They realized they "couldn't do that" because the disease was too infectious for them to do the whol-body blood transfusion they envisioned. They were, of course correct, no transfusion could do it...
So here is the "patch" for the movie to give them their ending, both visually and "technically" at least to the degree of "technically correct" established by the rest of the movie.
PATCH: The cure is basically the original virus that was engineered to become the super weapon. You give that to an uninfected donor person. Then as they go mad the original, slow way you give them a shot of bone marrow collected from the person you want to cure. This contains the infected immune cells of the intended recipient. The donor's immune system, primed with the more robust but slower original virus produces antibodies to unique viral strain created by the fast virus and the recipients genetics. The donor's blood serum now contains an antibody uniquely matched to the recipient which will jump-start their immune system. The donor cannot be re-used since introducing a second bone marrow sample would produce an immune complex that would simply kill the second recipient systemic organ rejection.
Okay, its all techno-babble, but it's _good_ techno-babble. It can be acted out on screen with big pantomime gestures. I has all the dramatic elements the writers and cinematographers were after. It actually makes more sense than a disease so virulent that it can drive you mad in seconds.
But that is the essence of a good patch, it changes the minimum while correcting the flaw and allowing for the desired outcome.
Apparently you have blotted out part of the movie to save yourself from the cognitive dissonance the horrible hidden planet engendered within your soul.
The planet was hidden. Go back and look at the scene when they are looking for the name Miranda and then looking at the empty orbital plot. How they puzzle about how there is nothing there. etc.
Since, by definition, it wasn't "in Reaver territory" before there were Reavers, there were dozens if not hundreds of years for people to go there when Miranda was just the outer-most planet in the system. Go there often. Perhaps weeky, delivering construction supplies. Inhabit it at least as well as they can inhabit empty space. Use it as a military base or a smugglers hide out. Use its gravity in slingshot maneuvers to get between the second and third outward most planets as they fulfilled the 500-year orbits dictated by the planets being far enough away from a main-sequence star so big it can afford a good dozens planets a stable orbit in its "habitable zone".
Or otherwise we hide this planet while the adventurous pioneering souls of spacefaring civilization all collectively forget the invention of the telescope and generally stop looking at the lights that move in the night sky.
And if you have ever hung out with people who go out to sea and make it back, were the place inherently bad prior to the Pax and the Reavers, then there would be drunken bar stories and warning beacons and all manner of record on the existence of the planet and its exact condition.
Not to even begin on the topic of all those dead people having families, heirs, and bank accounts; and some of them being nascient brown-coats or who went to college with investigative journalists; or who owed big money to that evil Russian Dude...
There is just _no_ excuse for Miranda.
Don laugh at Y2k. It was serious. Not the media part, but the actual washing of all that COBOL and PL/1 and all those terrible boundary cases. Much of the Y2k money was well spent.
But paying good salaries for no benefit, and DRM is _no_ benefit, sucks our money into a vast entropy sink. Sure some marketing execs, corporate execs, and charlatans skim a little off the flow of your cash into nothing. Notice how all those three profiteers have one thing in common? (okay, maybe several, but none of them are "smart producers of lasting value").
DRM is theft. An economic drag. A criminal dissipation of resources in pursuit of the idea the _totally_ _bizarre_ notion that there are a select few who deserve a lifetime of income for a days worth of work.
There, I said it. I think that copyright (the right to keep someone from bastardizing your work) should continue for the lifetime of an artist, but the charge-for-it-right ought to basically evaporate once everyone has made a decent wage for the time invested plus maybe 10% profit.
In what sane universe would I be paying Peter Frampton (deceased) or his heirs (all also deceased) for a performance of "Do you feel like I feel" if the purpose of copyright is, as stated, to encourage Peter to make more music (aparently from the beyond), while that "encouragement" happens by paying a record executive's lawyer to sue my dead grandmother?
Its all crazy talk. (and I am an author, so I understand the desire to keep the furries out of my mythos, but I don't see that the grand children of my nieces and nephews deserve to "live off of" me writing one book. [see Gone with the Wind.])
Coulda, sure, but the script and mythology already have habitable planets and moons, and even with a very, very, very wide/deep habitable zone, there is a minimum distance apart that 1-gravity-or-so planets would have to be separated by in order for them _neither_ become one large N-gravity planet _or_ pull each other into bits. After that they would have to be even further apart to keep them from giving each other gravity boosts into some very unlivable orbits.
So no, there is no excuse. Since their habitable planets have some habitable moons, you now have moons with moons of their own, all at about 1-gravity, orbiting a very huge and very distant sun so that you have a habitable zone as wide as the zone from earth to about neptune and about as far away from the central gravity well such that the nearest planet is about as far from the primary as (pulling a distance from my ass without any math at all 8-) Pluto is from Sol. (I make that totally wild guess just by trying straight scale-up the width of our habitable zone to carry seven plus planets of about earth normal gravity, Remembering that Venus is earth gravity or so but Mars is not, and both are marginally outside the prime habitable zone.)
That said, Making the Pax transmissible is super easy since I am editing the whoel ideal wholescale. It wasn't a "gas" it was a particulate aerosol introduced in the air handling system of a central facility (not the terraforming facilty plant) (or just sprayed from an airborne craft). In the field, possibly because of the idiosyncrasies we know every teraformed world developed (see "The Train Job") caused a prion disease (improperly folded protein). In most people the prion caused an amplified pacifying effect by destroying the aggression centers of the brain the original drug targeted. In some it destroyed the inhibition centers instead. The particulate is long gone, but the prions persist. Close contact with a reaver can infect you with the prion, especially if they maybe do the raping and maybe some of the eating, but then don't get around to the killing.
It's all fixable. Almost.
You just do _not_ get to "hide" a planet in the last minutes of the the last 1/5th or less of 100 or more years of planetary occupancy. If nothing else smugglers and criminals will occupy any gravity well that the Alliance doesn't. So a failed terraform would lead to a nice juicy "safe but abandoned" place, with a planet to hide behind and mine and whatnot, as resources and safety for, say, a brown-coat base during a civil war....?
Miranda, as presented in the movie, was just lame and was a total speed bump for my suspension of disbelief. I got over it to enjoy the rest of the movie.
Between the "one solar system" and "the secret planet" the movie took the mythos from "workable" to "cheap framework for space opera". Still loved the characters and the presentation, but they could have just called the whole place 'the forbidden zone' after pulling that much completely unnecessary mcguffen out of its ass.
I agree that Miranda would have been fine in a system where they weren't all orbiting one star.
And I would give you Reavers being the result of Pax if it were said that, perhaps, Pax was or caused a prion disease of the brain like Mad Cow disease etc. Easily done there.
Then again, the less-than-twenty (and likely less than ten) year timescale would belie Reavers being an old wives tale.
Draw out the time scale on paper sometime. Reavers would have been absolute fact, and a new fact at that, if the movie is cannon. They'd have exploded across the whole star system pervasively, and been hunted to extinction. They just wouldn't have the longevity (what with unshielded reactors and having to replenish their ranks and food supplies, since there isn't a 'reaver farmer' growing grain and we know food is scarce in that reality etc. In the absence of a reaver economy the would have been starved out of orbit at Miranda in like a month. etc.)
The whole thing reeks of vampire fan-boy logic. Hundereds of vampires inhabit 1840 Paris/Moscow/New York/New Orleans (as they have for hundreds of years etc), each one draining the blood from one human each night, year in and year out. So even in WWII where people die and go missing all the time 365.24 * 100 is, on average 36,524 people gone missing from one city each year. The numbers just don't work if you even glance sideways at the math.
To fix the problem with least effort: Keep the Pax, but try it on a cruise ship or a space station (e.g. "The Miranda" instead of "The Planet Miranda", or hell on a couple dozen people in a lab. Start with a smaller population. Make it properly contagious like the aforementioned prion disease (raver decides to rape you first, or feed you some of his flesh or blood, and _maybe_ you convert if you live long enough or get left behind) instead of a mystically repeatable experience that somehow has a predictable morphology, common enough for Mal to know the exact progression well enough to narrate it 8-). Same evil government, same evil plot, harder to come up with an armada of Ravers granted, but workable. Hell you even get to cut a word and some "look at the nothing" computer interface mock-up CGI from the script. Everybody wins. 8-)
Terraformign didn't take? Given that the war was there, and before the war there was a how many years or centuries for the aliance to grow up, grow dispised, grow dispotic, and engender warfare; well the system has been around a mighty long time. Everybody would have knows Miranda (the planet) was habitable long before the need to develop the Pax came about.
Hell, the firefly class being "good for smuggling" means that there was smuggling, which means that if Miranda (the planet) was ideally terraformed then everybody would have been there, and if it were a teraforming disaster, all the people who didn't want to be where all the people were woudl use it as a base just like they do everything else.
In the scale of things, having Miranda go unnoticed and discovered the way it was has the credibility of Atlantis. That is, it is as likely as finding out on the news this eventing that a land mass the size of Australia was just "discovered" _anywhere_ on the earth, and it had been hidden by the UN because 20 years ago they decided they needed a secret continent of their very own.
You just cannot hide a planet. And the _planet_ was hidden (watch the movie again, look at all that "nothing" in the CGI rendering of the not-the-internet when the discover the planet). yes, planet.
There just is no excusing that whole planet thing. Hell I came up with a way for the 28 Days Later people to pull off their originally preferred ending (watch the DVD extras) and that was _easy_ in that mythos. But outside of Buck Rodgers, circa 1930, you just _don't_ get license to hide a whole freaking planet co-orbiting a star occupied by multi-planetary spacefaring society full of smugglers and rebels.
Now, to be honest, when required to argue the other side in a different context, I took the position that the perturbations of navigation were hidden in the navigation system software. Nobody would have any cause to double-check the computer as long as the easy-to-computer compensations worked and everybody ended up at their destinations.
But I was defeated by the ten year old with binoculars. There is no way that a planetary mass, habitable and life sustaining at that, let alone one _surrounded_ by a cloud of reaver ships has an albedo low enough to hide, or even go unnoticed traversing the night sky of a few dozen planets and moons.
I love Joss' imagination. His vision is good. Like any bond villian, he needed to find him 5 year old to bounce his evil plan off of before putting it to paper. There is no excuse for Miranda.
Here were my ideas.
1) from the series, yep, some people go crazy when they spend too long on the edge of that much nothing.
2) in the series it was mentioned that, despite the rumors, there have never actually been any aliens or alien artifacts found. What if there were something so alien way, way out there, but attractive enough to draw people in, that was so alien that it broke the mind. (Would have dove-tailed in with River being prescient and mad, suppose the blue-hands were working from "the only reliable evidence" ever beamed back, trying to unlock the secrets and abilities of the thing without bringing on reaver madness.)
3) Evil government experimentation, version "not-dumb", something infectious. Call it Pax if you want. Give it the same reason and history, but instead of a "chemical" or in _addition_ to it being a chemical have raverdom be an Infectious Prion form of the original chemical (see mad cow disease). Still need to drop Miranda and the one central sun, but now being force-fed a little reaver flesh would maybe make you one of the family. Hell, "The Miranda" could have been a freaking cruise ship and the thing would work, but not a planet.
At a minimum, some serious retcon needs to take place to repair the plot damage the movie inflicted on the franchise.
I just wish they would ask me about this stuff before the script goes into production. 8-)
That and un-kill Wash and Sheppard Book.
Oh, and get rid of the whole Miranda bullshit. The people who ply the lanes of space would neither "overlook" nor "forget" an entire main planet over the course of less than 20 years. Nor could such a thing be hidden as, outer-most or not, it would show up on everybody's orbital computations as a huge perturbation in their plots. Let alone one ten-year-old with binoculars.
Oh yea, and drop that whole "all the planets orbiting one sun" nonsense since it isn't workable. Miranda would have been frozen ice-ball _or_ the "inner planets" would be molten slag.
Don't get me wrong, I loved the show. The movie needs to be declared out-of-cannon before the series would be workable.
I could have come up with a better "reason for the reavers" in my sleep. The original one from the series (mental erosion from facing the emptiness of space etc) was good enough. Hell, the movie contradicted the series directly. If the Pax caused reaverdom, the the episode where the one guy got tortured and became a reaver himself woudln't have worked unless the reavers carry a supply of the otherwise secret Pax around and deliberately pre-expose potential recruits to it before deciding who to kill, rape, and eat (in that order, if you're really lucky).
So yea, it needs a reboot.
DRM is the software version of Perpetual Motion. It is simply not possible to make the device described work as intended. But because of "the enormous potential income" should someone succeed, the greedy interest keep flushing money into the pockets of charlatans and charging the populace a tax for their stupid avarice.
Since DRM can only work if all the parts of the system are controlled by external DRM, including all the DRM enforcement parts, you end up with "its elephants all the way down."
So we will never be done until it is simply illegal. Just like the patent office will not accept patent applications for perpetual motion machines, and the FDA will not let unproved drugs out into the wild (in theory anyway 8-), the FTC (etc) will eventually need to refuse to let people try to sell things with this snake oil in it.
But like those remedies and limits, it will take a couple hundred years of corpses and bankruptcies cause by the offensive practice of duping companies into "DRM" before anybody finally acts to stop the scam.
And even then, people will still try to sneak it in the back door as "holistic systems engineering" or whatever.
You _aren't_ going to get a key for "The White Album" you are going to get a key for "the 2011 release of 'The White Album" in MP3 format from Sony Interactive for use on sPlayer #xxxxxxx" simply because they _can_ be that specific and they _don't_ want to sell anything once that they can sell a million times.
DRM == RENT, and illegal prior restraint, and a scheme that can never actually work because it is a system that violates every principle of both software engineering and cryptography. No matter how you slice it, DRM is a stupid waste of leptons, time, and money. It is a system based on a complete lack of modularity and locality.
DRM is a classic case of "who will watch the watchers?" and not just at the corporate and financial and cultural levels. As a simple exercise in software engineering DRM must fail. It is a system that must be part of every element of a system (which is the failure of locality and modularity etc) to the degree that you need to have DRM policing the DRM system.
DRM is the Perpetual Motion of Software. People keep inventing new versions of it that don't quite work because no version of it can _ever_ deliver what is promised. Companies keep buying into the hype because they are blinded by "the potential". The only difference is that we are all being forced to buy these perpetual motion machines. Sure _this_ one has a battery in it, _that_ one has to be hooked up to the electrical mains. Some other one needs a waterwheel or a solar panel, and they will all tear off an arm or crush your child if you aren't careful... but we are _almost_ there... just one more scheme and we'll have it right...
The whole thing is a tax, levied by the stupid, paid by the sheep, and ready to break businesses when, I don't know, say Microsoft (or whomever) forgets to update a certificate (or whatever) before it expires (or whatever).
Where the heck do I find the Opt-Out?
"If I deny, it don't apply." So you force your Assistant Mayor Equivalent to drink Cholera...
This is the kind of error messages you see from programmers who use exceptions _poorly_. I use exceptions well, and in doing so I don't just use the pointlessly empty exception classes. I have a tidy little toolbox with a slelect few exception classes that are normally instanced in a way that includes file and line numbers (for me) and useful text (for the user) at a minimum. Heck, when I compile with a debug macro defined (q.v. -DDEBUG) on a GCC/GlibC based system, my exception constructor saves a whole stack trace created at construction using backtrace().
Oddly enough, such exceptions are typically thrown in response to "doing error checking within the routine".
They are also damn handy inside of libraries where, for instance, you want to have a function that returns int and there is no "invalid" value to return (such as -1) because the whole domain of int is a valid value. (as in doing checksums or hash tables and so on).
The fact of the matter is, many people should not be trusted with power tools, and many people should not be trusted with manual tools, but nobody who thinks either kind of tool should not exist for whatever reason should be trusted with any tools at all.
You hate exceptions because either (a) you don't know how to use them properly or (b) you have been forced to use the code of someone who doesn't know how to use them properly, and perhaps (c) both of the above.
The real predicate is whether the condition can be properly handled locally or not. Passing an error return code (q.v. minus-one etc) back through ten layers of function calls is just as information poor a thing to do as throw an empty class as an exception. In fact its _worse_ in most cases as you need to make a tangle of logic to turn a callee's -1+errno for out-of-bounds into your local -1+errno for could-not-allocate, into the callers -1 pool-is-full-use-a-malloc-instead. This just multiplies for the full Cartesian Product of all possible error paths including those paths that also then fail.
Tell me true (presuming you use C/C++ on a posix-like os)...
Do you always check the return value from snprintf() to make sure that you didn't run out of buffer during conversion? Let me guess, you just choose to use really big buffers...
Do you wrap every write() in a pointer-increment loop to naturally catch when a write of ten bytes actually only writes 5. Let me guess, you only check for -1, and only on file descriptors that you think are "likely" to error out...
IMHO people who discard features whole-scale, and cite "hate" as a reason are generally guilty of practicing outside their ability.
There are plenty of things I "don't do", like I never use pure virtual functions in C++. I feel they serve no useful purpose except to annoy the programmer that comes along after me. (Having had a thread safety problem in a code block where one thread called a virtual function on a object that was being destroyed by another thread, and having the resultant call become a call to that NULL kinda irked me. But I learned my lesson. I put asserts or exceptions in the functions and document them, and document the class as having no useful default implementation. Also I've seen fat interfaces where the first thing a user has to do is stub-out the interface in order to make a single test call; where that lead to an important function ending up containing just "return 0;" all the way to production; which cost us time and money. But I don't "hate" pure virtual functions for finding them pointless and unhelpful.)
Er... Java is optional, and only required if you use the database engine (nobody does, because almost nobody knows when _not_ to use a spreadsheet, IMHO of course 8-) or some of accessibility and wizard thingies.
One of the ACs gives the actual quote and reference.
Plus OpenOffice.org (and I think core open office as well) dumped the larger desktop interface a long, long time ago.
Try something recent, and try reading the documentation, before you rail against any product.
Microsoft gets people to update by giving their product to the CEOs and "bigwigs". When everybody _else_ in the organization cannot read or use the new format for the documents, they have to keep bouncing transfered documents back to the aforementioned bigwigs. Eventually the bigwigs get tired of the fact that they cannot understand how to use save-as-older-format, and they dislike having their underlings telling them to do things, and they cannot bear to find all the files they saved and re-save them before they downgrade back to the old version... So the entire company naturally has to pay to upgrade everyone.
Repeat that at the border of the company. Every iteration of Little Company that works with and is dependent on Big Company, cannot allow themselves to be seen as unhelpful nor out of date, and they cannot bounce the documents they receive via email etc. without giving that exact impression...
Letting certificates expire is _not_ a Microsoft "strategy", it's an artifact of their adoption of "We don't care. We don't have to. We're The Phone Company" where there is no longer just one phone company, but Microsoft wants to be "The Software Company".
This _is_ egg on their face, but the only ones who will not yell "brilliant omelet" are the people who can connect the "Trusted Computing" dots. Letting the world _again_ see what it means to leave the keys to your property in the hands of any entity that doesn't _have_ to care is just another Microwhoops...