Comment On Flexible Spending

Frist.

Had to say that.

me no frist

Luckily the Garmin SaversGuide will guide me through the Untitled Polygon quite easily. All for a price that is significantly more than the US national debt.

To be frist your post needs to be between 0 and 0. Please try again.

Nobody noticed the counting error and started monopolizing this thread with it? I'm disappointed.

So, nancy in spain is hot?

I do not think there is enough currency in the world to purchase that SaversGuide Membership card

And I just noticed that it is only a 1.5 star rating.

Hey, the second-to-last one, at least they caught the exception! I recognize that error - it's thrown by the .net NumericUpDown control - and I've fixed a couple crashes-to-desktop that were the result of assuming there was no way invalid values would ever find their way to some control of that type somewhere, and thus not handling that in any way. (Noticed by me or a tester -before- releasing anything in that state, though.)

I actually live right down the street from Untitled Polygon. It really isn't a bad neighborhood at all.

Is that a Labview UI?

For shame...

There's an Untitled Polygon in Columbus? The entire state of Ohio should be renamed Untitled Polygon. After all, it's all about fancy places like California, New York, and China.

captcha: haero. I'm a Buckeye Haero.

MadScutter:

I actually live right down the street from Untitled Polygon. It really isn't a bad neighborhood at all.

Yeah, but you have to be careful because it's right next to Crouch End, which was built on top of a towen.

I don't see the WTF in the Carbon Copy Cloner screenshot. Obviously, the cloning process was a complete success, sou you now have to eject the disk twice - the original, and the clone.

The "Eject the disk, and then try to eject the disk again" phrase reminds me of a discussion I had about how many products aren't thoroughly tested in scenarios that involve the end of the product's life.

If you were building a starship (a la Star Trek), would you be willing to test the craft's self-destruct sequence all the way, or would you opt instead to shut it down a few seconds before the starship explodes?

Or would you even test it at all? For all you know, when you try to shut down the self-destruct sequence, you might get a friendly female computer voice advising, "In order to shut down the self-destruct sequence, you must first shut down the self-destruct sequence." That would spell disaster for a starship captain that figured out (literally at the last minute) a friendly, diplomatic solution to the current episode's problems.

NaN celsius? That´s why I stick to farenheit, i really cannot understand the metric system.

TRWTF is a neighborhood named "Milo Grogan." Maybe they should have left that polygon untitled, too.

J-L:

The "Eject the disk, and then try to eject the disk again" phrase reminds me of a discussion I had about how many products aren't thoroughly tested in scenarios that involve the end of the product's life.

If you were building a starship (a la Star Trek), would you be willing to test the craft's self-destruct sequence all the way, or would you opt instead to shut it down a few seconds before the starship explodes?

Or would you even test it at all? For all you know, when you try to shut down the self-destruct sequence, you might get a friendly female computer voice advising, "In order to shut down the self-destruct sequence, you must first shut down the self-destruct sequence." That would spell disaster for a starship captain that figured out (literally at the last minute) a friendly, diplomatic solution to the current episode's problems.

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

It is me.:

Frist.

Had to say that.

No you did not, unless there was a man with a gun pointed to the head of a beloved one and even though you don't have to say it.

I am working down the street form Untitled Polygon and honestly I really like the area.

captcha: odio - Garfields friend?

AP2:

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

But you don't have to build a complete starship; just the self-destruct unit and whatever explosives it uses. If the self-destruct is designed to cause a warp-core breach, that might have to be simulated with a test unit that samples the input and ensures that whatever commands it receives would actually cause the breach.

Yossi:

So, nancy in spain is hot?

Actually, I believe it's still winter there, so Nancy is probably pretty cold (NaN can come from calculations on negative values, too).

operagost:

AP2:

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

But you don't have to build a complete starship; just the self-destruct unit and whatever explosives it uses. If the self-destruct is designed to cause a warp-core breach, that might have to be simulated with a test unit that samples the input and ensures that whatever commands it receives would actually cause the breach.

I see where you are coming from AP2, but the examples you gave are not that great. Both planes and skyscrapers intended purpose is -not- to destroy itself as part of normal use, you need to be more specific, like how the seating in aircraft is design to destroy itself in the event of a crash (in order to absorb kinetic energy rather than you). As operagost pointed out, in those cases they test those emergency systems independently of the main unit and they -do- destroy a sample of them to make sure they work as design. This even applies to larger scale items like a self destruct device, they would perform test detonations of them before installing them in ships (in the case of Star Trek they probably would do some math to confirm that the explosion would be large enough to trigger a warp core breach). Even with our current nuke missiles they have been tested underground to make sure that they work, so if we ever did need them (which I hope is never) they would work. After all last thing you want is a dud in one of those missiles that would be one hell of an oops.

TGV:

Nobody noticed the counting error and started monopolizing this thread with it? I'm disappointed.

Yeah, I noticed it, but you beat me to the punch.

An off-by-one error in a TDWTF submission. Who woulda thunk it?

TRWTF is the eleventh message being 111.

Shame on you for trying to lure the unwitting masses into a KNOWN alien vortex! ;-)

The NaN in Spain stays mainly in the plain.

ThomasP:

TRWTF is the eleventh message being 111.

The eleventh message is probably not 111. TRWTF is the submitter stating that it is. Off by one. Let's hope he doesn't do C programming for a living.

Here in the Sahth, Nahn is ah pahfectly good nahmber. Comes between ahght and tahn.

Teamwork is great, right? Bring a diverse group of skills together and we all win, right?

So I'm in a project where I'm assigned all the tasks, because nobody else knows how to do them. But I have five people micromanaging every step (not counting my actual manager). And I have to use a large collection of spreadsheets so they can see what's going on, because they don't understand databases.

Bottom line this is going to take 10 times more work, plus salary for the five do-nothings, and even still, be more error prone than if I just wrote a bit of code against a database.

But at least we're working as a team! Yay team! Should get me high marks on my annual review. Giving other people a reason to exist.

Bananas:

TGV:

Nobody noticed the counting error and started monopolizing this thread with it? I'm disappointed.

Yeah, I noticed it, but you beat me to the punch.

An off-by-one error in a TDWTF submission. Who woulda thunk it?

So Trevor is TRWTF?

EDIT: Apparently, a couple of you beat me to it. TGV, are you happy now?

Mason Wheeler:

The NaN in Spain stays mainly in the plain.

Or plane.

If you want Amazon insanity, just look up the tag wtf. It's hard to beat the Example of one of my images on the side of a truck from $1,294,117,630,223,910.00.

So the MCP guy managed to not provide a first name when creating his Passport/Live ID/whatever-it's-called-now and he's surprised that the prompt Not <empty string>? shows up? So yeah, Microsoft ought to have some default text, like maybe the email address instead of the name. Not the highest on the WTF scale.

frits:

Is that a Labview UI?

For shame...

Yep, definitely LabVIEW. I bet if we could see the "code" behind that interface it would easily be TRWTF of this article.

John:

Mason Wheeler:

The NaN in Spain stays mainly in the plain.

Or plane.

I'm sick of these motherfucking NaNs on this motherfucking plane! (or plain...?)

AP2:

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

Goodness me, you are so naive. What you do is, you design these things with redundant subsystems, each of which can be tested individually and with a view to their interoperability. You test these subsystems to destruction according to Anketam's posting. You design a fill-scale "testability" mode, with (as has been suggested earlier) a specific mode to test the self-destruct mechanism, which *will* go the full way but just not actually firing off the explosives themselves, so as to ensure that the process works correctly. Now *do* try to catch up.

Maurits:

Here in the Sahth, Nahn is ah pahfectly good nahmber. Comes between aiyught and taiyuhn.

Fixed thaiyuht fur yew.

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That's not how QA works at all (or confidence intervals for that matter).

Are you me, posting from an only slightly alternate universe? Because that sure sounds like what I have to put up with,

PedanticCurmudgeon:

Bananas:

TGV:

Nobody noticed the counting error and started monopolizing this thread with it? I'm disappointed.

Yeah, I noticed it, but you beat me to the punch.

An off-by-one error in a TDWTF submission. Who woulda thunk it?

So Trevor is TRWTF?

EDIT: Apparently, a couple of you beat me to it. TGV, are you happy now?

First thing I noticed about the whole set of submissions (other than a remote possibility of referring to Spanish NanC-boy weather). Only reason I wasn't the first to post was that I slept late this morning.

Valencia is completely broke.

I find the Windows icon intriguing. So compressing a hard disk is like sticking two oversized band aids on your monitor?

I don't believe that Greg D knows the Untitled Polygon neighborhood at all, as there's no reason to steer clear of it. It's an upper-middle class area with a really low crime rate, and is actually one of the nicer areas of Columbus.

http://www.city-data.com/neighborhood/Untitled-Polygon-Columbus-OH.html

Certified Microsoft professional. LoL

- Hey Bob I got a problem with my kitchen.
- Have you tried upgrading your windows license?

operagost:

AP2:

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

But you don't have to build a complete starship; just the self-destruct unit and whatever explosives it uses. If the self-destruct is designed to cause a warp-core breach, that might have to be simulated with a test unit that samples the input and ensures that whatever commands it receives would actually cause the breach.

Sure. A full end-to-end test is something you can only do a limited number of times. Realistically you do not build a brand new starship and use it to test the self-destruct. You do like the US did to test early nuclear weapons: get some old,obsolete ships that were going to be scrapped anyway, and test on them.

You can, of course, test the explosive in the middle of the desert or in deep space or whatever with no star ship attached. This would test that the firing mechanism works, it really does detonate, etc. That would leave you with the problem of being sure that the explosion would actually destroy the starship. Or, I suppose, that there is not some subtle problem caused by installing it in the starship, like you didn't realize that the wiring goes past the transporter room where the neutrino flux interferes with the signal. But a wise engineer would test that sort of thing by having putting test equipment where the explosives would be and insuring that they get the proper signal.

Of course, the engineers on Star Trek all seemed to be absolute idiots. Like, remember the episode "The Ultimate Computer"? They build a computer that is supposed to control a starship in a combat situation. And for what is apparently the first test, they put it on a ship with real weapons and provide no means to override or disable it. So when the system fails, it starts killing people and there's no way to stop it. Umm, wouldn't any engineer who was setting up a test of a device that controls weapons capable of literally destroying a planet, and who was not a complete idiot, include an "off" switch just in case the system fails? (Of course if they had done that, the episode would have been very short: "Captain, M5 is powering up the phasors to attack an unarmed freighter!" "Yikes, we'd better turn it off and head back to base." "Okay, done. Sorry Dr Whatever, your computer doesn't work. You'll need to work out the bugs there." The end.)

Medezark:

Are you me, posting from an only slightly alternate universe? Because that sure sounds like what I have to put up with,

No, I'm not you, unless they don't have "Quote" or "Preview" buttons in your universe.

But still, good to know others are suffering along with me. I think.

Hey, as you brought up Star Trek's self-destruct mechanism, something about the movie where they use it really baffled me.

A small alien ship attacks the Enterprise. Kirk cleverly fools them into sending a boarding party that appears to be about six people. He and the crew then escape after setting the self-destruct, thus killing the boarding party. He and the crew then celebrate how this brilliant ploy resulted in their defeating the enemy.

But, umm, the Enterprise is supposed to be one of the biggest, most powerful ships in the fleet. Comparable to, say, an aircraft carrier in a 21st century navy.

Suppose that a band of Somali pirates attacked a U.S. aircraft carrier. So the captain evacuates the crew from the ship and lures six of the pirates on board, and then he blows up and sinks the aircraft carrier, taking the six pirates down with it. He then reports back to the Pentagon, "Admiral, through a clever trick, we managed to kill six Somali pirates, at the cost of only one multi-billion dollar aircraft carrier!" Do you think he'd get a medal for that? Or would they be amazed at the incredible idiocy of sacrificing one of the most powerful and expensive ships in the fleet to kill a mere SIX enemy soldiers?

AP2:

Anketam:

There is already a methodology for testing this, it is called QA. This scenario plays itself in many other fields which involve reliability of items that destroy themselves when used, like emergency flares, rockets, emergeny rafts, and more. How they test them is by taking a meaningful sample and testing them all the way. Assuming the testing goes well then QA will come back and say something to the affect that: "We can say with a 95% confidence that said item will function as designed 99.9% of the time (or whatever the target is)." Can item still fail at the most critical moment, yes, but it has been proven statisticly to be negligible.

That only works for cheap mass produced products, which is not the case the example given (the Star Trek Enterprise), of which only 8 were built, or similar constructions. If you're building a skyscraper or a transcontinental plane, you can't build hundreds of them just so you have a statistically valid sample to test.

They did so in Malaysia ... and the self-destruct test failed.

PedanticCurmudgeon:

Apparently, a couple of you beat me to it. TGV, are you happy now?

Not until the majority of the posts derives from one of these posts, but goes off on a few pedantic personal attacks and attempts to relate the topic to women.

frits:

That's not how QA works at all (or confidence intervals for that matter).

General disclaimer about my posts: I at no point claim to be a subject matter expert on anything, as such I can say something that is wrong (this is what Google is for).

Jay:

Hey, as you brought up Star Trek's self-destruct mechanism, something about the movie where they use it really baffled me.

A small alien ship attacks the Enterprise. Kirk cleverly fools them into sending a boarding party that appears to be about six people. He and the crew then escape after setting the self-destruct, thus killing the boarding party. He and the crew then celebrate how this brilliant ploy resulted in their defeating the enemy.

But, umm, the Enterprise is supposed to be one of the biggest, most powerful ships in the fleet. Comparable to, say, an aircraft carrier in a 21st century navy.

Suppose that a band of Somali pirates attacked a U.S. aircraft carrier. So the captain evacuates the crew from the ship and lures six of the pirates on board, and then he blows up and sinks the aircraft carrier, taking the six pirates down with it. He then reports back to the Pentagon, "Admiral, through a clever trick, we managed to kill six Somali pirates, at the cost of only one multi-billion dollar aircraft carrier!" Do you think he'd get a medal for that? Or would they be amazed at the incredible idiocy of sacrificing one of the most powerful and expensive ships in the fleet to kill a mere SIX enemy soldiers?

Note it has been many years since I have seen it, so if someone knows better please correct me. The ship in question was going to be mothballed so it was not that big of a loss, they only had a skeleton crew of like 5 people so there was no way that they could hold the ship even against a small boarding party.