In defense of said muppet. This has got to be a case of poor code commentation causing a maintenance problem.
The muppet was responding to a warning put out by an analysis tool (Valgrind) that some uninitialized memory was being read from, any programmer will tell you that's not a fantastic idea in general but it happens that in this case it was not only ok, but necessary.
But the muppet came along and "fixed" the problem.
Now this can only mean that there was inadequate code commenting (or, that the muppet didn't read it, in which case I retract my defense, but not the reasoning) around this unusual piece of code. If a piece if code is doing something decidedly strange, they there NEEDS to be a comment regarding it, preferably informing the reader of what is happening, or at least saying here be dragons or similar.
All it needed was:/* Reading some uninitialized memory to get a randomization seed. */
Many linux distros claim to be "easy", and they are, if you are at least moderately computer literate.
What I am looking for however is a truly Gran friendly Linux distro, for the quite elderly (who have done all the learning they will be doing, and will forget everything you teach them in about 1/2 an hour anyway and the just terminally stupid. Ultra, mega, hyper simple.
A desktop of about 5 buttons
"Mail" "Search" "Chat" "Write" "Pictures"
Only one way to do anything, a simple way, in fact, one mouse button. No guess work.
Like a mac, only even more restricted, locked down, and simple.
Mod that man up! There is NOTHING wrong with a paper and pencil census. In fact, it's probably the cheapest and most efficient method.
I live in New Zealand, now admittedly we are a very small country with only around 4 million people, but our recent paper and pencil census worked just fine. For a few weeks a bunch of temporary staff went to every house in the country and hand delivered sufficient census forms for the people living in said house On census night (the census is to be a snapshot of the population on a specific night) everybody filled out their census forms. And then the census people visited every house again and picked up the forms.
Once they get back to the statistics office they get run through the reading machine.
It's not like voting, you don't need the results quickly, it could take a year to collate the results if necessary. It would be such a long drawn out process if the census takers had to go around to every PERSON and take this 20 or 30 minute survey with each and every one, it wouldn't be a snapshot on a single day, it'd lose effectiveness.
As for people complaining there is too much information asked... well, sorry but if you want your government to provide the services which people like you need, then they need to know some stuff about you all and the census is the best way to do that.
Pah. The first (and last) time I watched one of these "battle bots" programmes, I was left thinking "where are the robots". All I saw was (somewhat dangerous) remote controlled toys.
When your battle bot can battle even somewhat autonomously, then I'll watch.
Apple is less about choice and more about giving them the best experience possible.
To an extent, but it's borne by giving the most homogenous experience possible.
Apple is all about "the one true way".
I'm not saying that's a bad thing at all, infact I fully support the ideal, removing complexity, removing choices, just give the user one (good, simple) way to do something, lose all the other ways to skin the proverbial. KISS.
Your point raises an interesting question. What is Google doing with Google Video?
Google Video offers much higher quality, longer videos, allows pay-per-view type deals etc, much more suitable for actual useful videos, and yet, Google appears to be, well, not pushing it now that they have YouTube.
If I was Google, I'd be combining the two, bring the features of Google Video to YouTube, or vice versa. But it doesn't appear to be happening.
"It's an undocumented feature. We're not going to document it because we don't want to support it. Feel free to use it, but don't bother complaining when it breaks sometime in the future."
Re:Note that Mapmakers make intentional mistakes..
on
Open US GPS Data?
·
· Score: 1
I can attest to this because near where my parents live on most maps there is a road that appears to go from their development right into the next one. Unfortunately, there is a gap of about 100 feet where there is no road but rather a swampy stream.
This example doubtlessly has nothing to do with copyright protection. It's simply a "paper road". Nothing sinister in that, there are probably millions of paper roads in the world which either were once in the past, will be in the future, or may never even become, proper formed roads.
[ Ahh fuck. How did I do that. Changed it to POT mode before I submitted. Habit. Let's try that again. Wish/. would allow editing posts Digg style. Filling in time because of the annoying required delay. Using extrans mode, so I'm quoting old school style here... ]
Qbertino> There is no point bolting another layer of that on top of them.
Hear hear!
All these freaking template languages just re-invent the wheel for the most part. PHP *IS* a template language first and foremost, that's what it was designed to do, that's why we can intermingle PHP with other stuff in the first place. What the hell is wrong with
<?php foreach($x as $y) { ?>... <?php echo $x ?>... <? } ?> how is it any worse than
{loop y as x} {echo x} {/loop} and other such constructs which I've seen in "template languages" which invariably do nothing except slow down...everything.
Qbertino> anything other that TAL (Pe(rl)-TAL, PHP-TAL or 'original' TAL
I'll have to disagree here, I just looked at PHP-TAL and it looks little better than anything else I've seen.
The only things I want out of any template language is things to automate stuff. If it doesn't make something EASIER for me, or for designers, or for.. somebody, then I don't want it, it's useless cycle-wasting junk.
{$Foo} instead of <?php echo htmlspecialchars($Foo) ?> is good, provided you can use both. <form:input name="MyForm[Blah]" validate="something" type="text"/> is good instead of a form input and a whole bunch of php/javascript to prefil the value, remember it between posts, perform validation and error checking etc.
But when a template language re-invents loops (tal:repeat="item itemsArray") then I have to say STOP that's retarded, PHP can loop for you JUST FINE.
Now all this said. I was guilty once of writing a template language that re-invented loops, conditionals and other such constructs. The reason for doing so was a) proof of concept, b) I was young and stupid, c) I thought it would make it easier for designers using Dreamweaver. I used it a couple of times, and then thankfully saw how RETARDED I was, how it didn't make ANYTHING easier for anybody, infact it made it more difficult, and promptly threw all that completely pointless crap away.
[ Using extrans mode, so I'm quoting old school style here... ]
Qbertino> There is no point bolting another layer of that on top of them.
Hear hear!
All these freaking template languages just re-invent the wheel for the most part. PHP *IS* a template language first and foremost, that's what it was designed to do, that's why we can intermingle PHP with other stuff in the first place. What the hell is wrong with...... how is it any worse than
{loop y as x} {echo x} {/loop} and other such constructs which I've seen in "template languages" which invariably do nothing except slow down...everything.
Qbertino> anything other that TAL (Pe(rl)-TAL, PHP-TAL or 'original' TAL
I'll have to disagree here, I just looked at PHP-TAL and it looks little better than anything else I've seen.
The only things I want out of any template language is things to automate stuff. If it doesn't make something EASIER for me, or for designers, or for.. somebody, then I don't want it, it's useless cycle-wasting junk.
{$Foo} instead of is good, provided you can use both. is good instead of a form input and a whole bunch of php/javascript to prefil the value, remember it between posts, perform validation and error checking etc.
But when a template language re-invents loops (tal:repeat="item itemsArray") then I have to say STOP that's retarded, PHP can loop for you JUST FINE.
Now all this said. I was guilty once of writing a template language that re-invented loops, conditionals and other such constructs. The reason for doing so was a) proof of concept, b) I was young and stupid, c) I thought it would make it easier for designers using Dreamweaver. I used it a couple of times, and then thankfully saw how RETARDED I was, how it didn't make ANYTHING easier for anybody, infact it made it more difficult, and promptly threw all that completely pointless crap away.
You know, I'm apparently INTJ. I think it's a type fairly common here on slashdot.
On/. yes INTJs form an unusually high proportion (for INTJs) of the audience. Global population speaking INTJs form a very VERY small proportion of the total. The INTJ personality type description will tell you why that is.
However population size of the INTJ does not factor into why we get small numbers of contacts (since INTJs are not necessarily, or advisedly, looking to hook up with INTJs), INTJs are simply by their nature, less interesting to potential mates, less interesting than even other Introverted types.
Include in this the factor that I live in an populationarily sparse island of a country which is about as geographically far away from any other country as can be, and the country has a total population of about 4 million which in general has a fixation on rugby, racing and beer (in none of which I personally partake), and a handful of contacts a year is actually doing bloody well.
In fact, this whole reply is a pretty INTJ thing to write, the majority of the population out there would just look at this and think "can this guy be any more boring".
Trader A: I will pay the holder of this option $80/share, if you present the option to me before the 1st of January.... time passes... Trader B: It's the 28th of December, and the share price is still $100/share, this option is worthless. Trader C: Hey, I like worthless things, I'll pay you $1/share to take those options off your hands....30th of December... Company: Uh oh, we tanked. Trader B: ARGH! Who will buy this junk stock, going cheap, $20/share! Trader C: Hey, I'm a nice guy, I'll buy that. Trader C: Hey, Trader A, look what I got here, please cough up your $80/share for this stock that I just bought for $20/share!
1. Generate XML 2. Parse XML 3. Encode Binary from Parsed XML 4. Transmit Binary 5. Decode Binary 6. Generate XML from Decoded Binary 7. Parse XMl 8. Actually do something useful.
e. So if you purchased something legally and you have no reason to think it might be stolen, then usually the worst that happens is that the original owns has to offer you a fair market price for the return of the item.
In what country is that? As far as I'm aware, in most countries, if you buy something which was stolen, if you knew it was or not, the original owner gets it back and you get nothing other than to claim compensation from the person who sold you the stolen goods. If you were aware, then not only are you out of luck, but you're also either in jail or looking at a very hefty fine.
Now you want to fix that bug without breaking pages(*).
THIS is the problem. How on earth did we get into this mindset that every browser revision must not cause "breakage" (for want of a better word) to existing pages which were supposed to have been adherant to a documented standard by virtue of containing an appropriate doctype.
Those pages that "break" were broken before, it's up to the person responsible for those pages to fix them by testing thier site in that new browser and seeing if anything needs doing (in the pre-release period if thier smart).
I wouldn't mind if browsers implemented a meta-tag/http header check which said "render this as an older non compliant version", it just makes the "fixing" job easier for those people who wrote broken code because they only need to add this tag. But I certainly do NOT support a tag that effectively says "render this with the best adherance to the standards you can, I know I already told you that in the doctype, but I really mean it this time, really!".
I blame designers largely, partiularly those ones (and sadly, in my experience, it's the majority) who get pissed off when thier site doesn't look EXACTLY (to the pixel) the same in all the different browsers. Not only are they likely to employ various hackery to get it that way, but they are also the ones to jump up and down when a new browser comes out and makes thier design shift a pixel to the right.
Banks do create the money, for the simplistic explanation, an example:
You have $100. I have a bank, it has no money. Alice wants some money, and has none. You deposit your $100 in my bank. I loan Alice $90. How much money do you have - $100, even if not physically, but it's still yours, you can still spend it at least electronically. How much money does Alice have - $90. So we have $190 now, $90 was created by the Bank.
It gets worse. Alice gives her $90 to Bob to buy a house. Bob puts the $90 in my bank. I keep $10 in reserve and loan $80 to Jack. So now from that original $100, you have $100, Bob has $90 and Jack has $80.
That is (very simplistically) how banks create the majority of the money in the system. The actual printing of money is a tiny drop in the bucket.
So then, what? Create/edit a file on your working drive, save it there, quit the app, and drag the file to the backup drive via Windows?
Well, the copying to the backup should be automatic, but yes, this is exactly what should happen.
It seems that you don't quite grasp the idea of a backup, the general plan is that you work on working files, and these are duplicated at some point in time to a separate location. The only time you should ever even look at the backup set is when recovering files (or testing your recovery process), and you should NEVER be editing backups, because then your backup is not a backup any longer but a working set!
No, he just showed that sadly like the majority of developers, he is unable to distinguish between LANGUAGE and API, at least when it comes to javascript (language) and dom (api) - it is evident all the time when people cry that they hate javascript because they can't get X bit of dynamic html to look right in all browsers, they don't see or understand that Javascript has nothing to do with that, it's the API, the DOM that is causing the grief.
As you say, when it comes to speed, especially with something like JS+DOM it's not the language that's the problem because the language isn't actually doing a hell of a lot unless you're going to be doing some major number crunching or something, it's the api calls that take the time (in application programming land, the syscalls).
Any benchmark of JS+DOM needs to weigh heavily on the DOM part of that pair to be of any real help.
Parachute is pointless in almost all modern aircraft for two reasons...
a) Likelyhood of aircraft actually catasrophically failing (breaking up) at a height at which parachute would be of any help is almost nil, almost all accidents were not accidents until the perfectly flyable plane hit the ground. b) Likelyhood of occupants of a catasprophically failed aircraft, which is now falling to earth, actually getting out of said aircraft, let alone deploying chute at sufficient height, is essentially nil.
If you're flying a one or two seat aircraft aerobatically, or in combat, or gliding (which involves lots of more or nearly aerobatic manouvering) where losing a wing or control surfaces at altitude is a possibilty, and getting out not too problematic, then wear a chute. For everybody else, it's just dead weight.
Being a pragmatist, I just don't see the point in Microsoft spending so much time, money and energy developing thier own engine when they can just take another one, write a legacy wrapper and a few patches to maintain compatability with old stuff, deprecate that old stuff, and move on with life.
Slashdot Mirror
In defense of said muppet. This has got to be a case of poor code commentation causing a maintenance problem.
The muppet was responding to a warning put out by an analysis tool (Valgrind) that some uninitialized memory was being read from, any programmer will tell you that's not a fantastic idea in general but it happens that in this case it was not only ok, but necessary.
But the muppet came along and "fixed" the problem.
Now this can only mean that there was inadequate code commenting (or, that the muppet didn't read it, in which case I retract my defense, but not the reasoning) around this unusual piece of code. If a piece if code is doing something decidedly strange, they there NEEDS to be a comment regarding it, preferably informing the reader of what is happening, or at least saying here be dragons or similar.
All it needed was: /* Reading some uninitialized memory to get a randomization seed. */
Many linux distros claim to be "easy", and they are, if you are at least moderately computer literate. What I am looking for however is a truly Gran friendly Linux distro, for the quite elderly (who have done all the learning they will be doing, and will forget everything you teach them in about 1/2 an hour anyway and the just terminally stupid. Ultra, mega, hyper simple. A desktop of about 5 buttons "Mail" "Search" "Chat" "Write" "Pictures" Only one way to do anything, a simple way, in fact, one mouse button. No guess work. Like a mac, only even more restricted, locked down, and simple.
Paedofinder General sums it up... http://www.youtube.com/watch?v=jaUkt59vY1Q
Mod that man up! There is NOTHING wrong with a paper and pencil census. In fact, it's probably the cheapest and most efficient method.
I live in New Zealand, now admittedly we are a very small country with only around 4 million people, but our recent paper and pencil census worked just fine. For a few weeks a bunch of temporary staff went to every house in the country and hand delivered sufficient census forms for the people living in said house On census night (the census is to be a snapshot of the population on a specific night) everybody filled out their census forms. And then the census people visited every house again and picked up the forms.
Once they get back to the statistics office they get run through the reading machine.
It's not like voting, you don't need the results quickly, it could take a year to collate the results if necessary. It would be such a long drawn out process if the census takers had to go around to every PERSON and take this 20 or 30 minute survey with each and every one, it wouldn't be a snapshot on a single day, it'd lose effectiveness.
As for people complaining there is too much information asked... well, sorry but if you want your government to provide the services which people like you need, then they need to know some stuff about you all and the census is the best way to do that.
Pah. The first (and last) time I watched one of these "battle bots" programmes, I was left thinking "where are the robots". All I saw was (somewhat dangerous) remote controlled toys.
When your battle bot can battle even somewhat autonomously, then I'll watch.
To an extent, but it's borne by giving the most homogenous experience possible.
Apple is all about "the one true way".
I'm not saying that's a bad thing at all, infact I fully support the ideal, removing complexity, removing choices, just give the user one (good, simple) way to do something, lose all the other ways to skin the proverbial. KISS.
Your point raises an interesting question. What is Google doing with Google Video?
Google Video offers much higher quality, longer videos, allows pay-per-view type deals etc, much more suitable for actual useful videos, and yet, Google appears to be, well, not pushing it now that they have YouTube.
If I was Google, I'd be combining the two, bring the features of Google Video to YouTube, or vice versa. But it doesn't appear to be happening.
I'd hate to see Google Video bite the dust.
No all they need to say is
"It's an undocumented feature. We're not going to document it because we don't want to support it. Feel free to use it, but don't bother complaining when it breaks sometime in the future."
This example doubtlessly has nothing to do with copyright protection. It's simply a "paper road". Nothing sinister in that, there are probably millions of paper roads in the world which either were once in the past, will be in the future, or may never even become, proper formed roads.
[ Ahh fuck. How did I do that. Changed it to POT mode before I submitted. Habit. Let's try that again. Wish /. would allow editing posts Digg style. Filling in time because of the annoying required delay. Using extrans mode, so I'm quoting old school style here... ]
... <?php echo $x ?> ... <? } ?>
/> is good instead of a form input and a whole bunch of php/javascript to prefil the value, remember it between posts, perform validation and error checking etc.
Qbertino> There is no point bolting another layer of that on top of them.
Hear hear!
All these freaking template languages just re-invent the wheel for the most part. PHP *IS* a template language first and foremost, that's what it was designed to do, that's why we can intermingle PHP with other stuff in the first place. What the hell is wrong with
<?php foreach($x as $y) { ?>
how is it any worse than
{loop y as x} {echo x} {/loop}
and other such constructs which I've seen in "template languages" which invariably do nothing except slow down...everything.
Qbertino> anything other that TAL (Pe(rl)-TAL, PHP-TAL or 'original' TAL
I'll have to disagree here, I just looked at PHP-TAL and it looks little better than anything else I've seen.
The only things I want out of any template language is things to automate stuff. If it doesn't make something EASIER for me, or for designers, or for.. somebody, then I don't want it, it's useless cycle-wasting junk.
{$Foo} instead of <?php echo htmlspecialchars($Foo) ?> is good, provided you can use both. <form:input name="MyForm[Blah]" validate="something" type="text"
But when a template language re-invents loops (tal:repeat="item itemsArray") then I have to say STOP that's retarded, PHP can loop for you JUST FINE.
Now all this said. I was guilty once of writing a template language that re-invented loops, conditionals and other such constructs. The reason for doing so was a) proof of concept, b) I was young and stupid, c) I thought it would make it easier for designers using Dreamweaver. I used it a couple of times, and then thankfully saw how RETARDED I was, how it didn't make ANYTHING easier for anybody, infact it made it more difficult, and promptly threw all that completely pointless crap away.
[ Using extrans mode, so I'm quoting old school style here... ]
... ...
Qbertino> There is no point bolting another layer of that on top of them.
Hear hear!
All these freaking template languages just re-invent the wheel for the most part. PHP *IS* a template language first and foremost, that's what it was designed to do, that's why we can intermingle PHP with other stuff in the first place. What the hell is wrong with
how is it any worse than
{loop y as x} {echo x} {/loop}
and other such constructs which I've seen in "template languages" which invariably do nothing except slow down...everything.
Qbertino> anything other that TAL (Pe(rl)-TAL, PHP-TAL or 'original' TAL
I'll have to disagree here, I just looked at PHP-TAL and it looks little better than anything else I've seen.
The only things I want out of any template language is things to automate stuff. If it doesn't make something EASIER for me, or for designers, or for.. somebody, then I don't want it, it's useless cycle-wasting junk.
{$Foo} instead of is good, provided you can use both. is good instead of a form input and a whole bunch of php/javascript to prefil the value, remember it between posts, perform validation and error checking etc.
But when a template language re-invents loops (tal:repeat="item itemsArray") then I have to say STOP that's retarded, PHP can loop for you JUST FINE.
Now all this said. I was guilty once of writing a template language that re-invented loops, conditionals and other such constructs. The reason for doing so was a) proof of concept, b) I was young and stupid, c) I thought it would make it easier for designers using Dreamweaver. I used it a couple of times, and then thankfully saw how RETARDED I was, how it didn't make ANYTHING easier for anybody, infact it made it more difficult, and promptly threw all that completely pointless crap away.
On
However population size of the INTJ does not factor into why we get small numbers of contacts (since INTJs are not necessarily, or advisedly, looking to hook up with INTJs), INTJs are simply by their nature, less interesting to potential mates, less interesting than even other Introverted types.
Include in this the factor that I live in an populationarily sparse island of a country which is about as geographically far away from any other country as can be, and the country has a total population of about 4 million which in general has a fixation on rugby, racing and beer (in none of which I personally partake), and a handful of contacts a year is actually doing bloody well.
In fact, this whole reply is a pretty INTJ thing to write, the majority of the population out there would just look at this and think "can this guy be any more boring".
Damn this INTJness, I only get 3-4 messages on a good YEAR!
To reduce your description a little further:
... time passes ... ...30th of December...
Trader A: I will pay the holder of this option $80/share, if you present the option to me before the 1st of January.
Trader B: It's the 28th of December, and the share price is still $100/share, this option is worthless.
Trader C: Hey, I like worthless things, I'll pay you $1/share to take those options off your hands.
Company: Uh oh, we tanked.
Trader B: ARGH! Who will buy this junk stock, going cheap, $20/share!
Trader C: Hey, I'm a nice guy, I'll buy that.
Trader C: Hey, Trader A, look what I got here, please cough up your $80/share for this stock that I just bought for $20/share!
1. Generate XML
2. Parse XML
3. Encode Binary from Parsed XML
4. Transmit Binary
5. Decode Binary
6. Generate XML from Decoded Binary
7. Parse XMl
8. Actually do something useful.
Yay for efficiency.
Use svnmerge[.py] - it's still almost slower than the pitch drop experiment, but it does do the job.
In what country is that? As far as I'm aware, in most countries, if you buy something which was stolen, if you knew it was or not, the original owner gets it back and you get nothing other than to claim compensation from the person who sold you the stolen goods. If you were aware, then not only are you out of luck, but you're also either in jail or looking at a very hefty fine.
THIS is the problem. How on earth did we get into this mindset that every browser revision must not cause "breakage" (for want of a better word) to existing pages which were supposed to have been adherant to a documented standard by virtue of containing an appropriate doctype.
Those pages that "break" were broken before, it's up to the person responsible for those pages to fix them by testing thier site in that new browser and seeing if anything needs doing (in the pre-release period if thier smart).
I wouldn't mind if browsers implemented a meta-tag/http header check which said "render this as an older non compliant version", it just makes the "fixing" job easier for those people who wrote broken code because they only need to add this tag. But I certainly do NOT support a tag that effectively says "render this with the best adherance to the standards you can, I know I already told you that in the doctype, but I really mean it this time, really!".
I blame designers largely, partiularly those ones (and sadly, in my experience, it's the majority) who get pissed off when thier site doesn't look EXACTLY (to the pixel) the same in all the different browsers. Not only are they likely to employ various hackery to get it that way, but they are also the ones to jump up and down when a new browser comes out and makes thier design shift a pixel to the right.
Banks do create the money, for the simplistic explanation, an example:
You have $100.
I have a bank, it has no money.
Alice wants some money, and has none.
You deposit your $100 in my bank.
I loan Alice $90.
How much money do you have - $100, even if not physically, but it's still yours, you can still spend it at least electronically.
How much money does Alice have - $90.
So we have $190 now, $90 was created by the Bank.
It gets worse.
Alice gives her $90 to Bob to buy a house.
Bob puts the $90 in my bank.
I keep $10 in reserve and loan $80 to Jack.
So now from that original $100, you have $100, Bob has $90 and Jack has $80.
That is (very simplistically) how banks create the majority of the money in the system. The actual printing of money is a tiny drop in the bucket.
For the far more interesting story, I can highly recommend viewing the first video here, it's really worth it:
http://video.google.com/videosearch?q=money+is+debt&sitesearch=
Well, the copying to the backup should be automatic, but yes, this is exactly what should happen.
It seems that you don't quite grasp the idea of a backup, the general plan is that you work on working files, and these are duplicated at some point in time to a separate location. The only time you should ever even look at the backup set is when recovering files (or testing your recovery process), and you should NEVER be editing backups, because then your backup is not a backup any longer but a working set!
No, he just showed that sadly like the majority of developers, he is unable to distinguish between LANGUAGE and API, at least when it comes to javascript (language) and dom (api) - it is evident all the time when people cry that they hate javascript because they can't get X bit of dynamic html to look right in all browsers, they don't see or understand that Javascript has nothing to do with that, it's the API, the DOM that is causing the grief.
As you say, when it comes to speed, especially with something like JS+DOM it's not the language that's the problem because the language isn't actually doing a hell of a lot unless you're going to be doing some major number crunching or something, it's the api calls that take the time (in application programming land, the syscalls).
Any benchmark of JS+DOM needs to weigh heavily on the DOM part of that pair to be of any real help.
Parachute is pointless in almost all modern aircraft for two reasons...
a) Likelyhood of aircraft actually catasrophically failing (breaking up) at a height at which parachute would be of any help is almost nil, almost all accidents were not accidents until the perfectly flyable plane hit the ground.
b) Likelyhood of occupants of a catasprophically failed aircraft, which is now falling to earth, actually getting out of said aircraft, let alone deploying chute at sufficient height, is essentially nil.
If you're flying a one or two seat aircraft aerobatically, or in combat, or gliding (which involves lots of more or nearly aerobatic manouvering) where losing a wing or control surfaces at altitude is a possibilty, and getting out not too problematic, then wear a chute. For everybody else, it's just dead weight.
If I was running the IE8 team, I'd be a bit more ruthless..
Step 1: Replace Trident with Gecko/Opera/WebKit
Step 2: ???
Step 3: Profit!
Being a pragmatist, I just don't see the point in Microsoft spending so much time, money and energy developing thier own engine when they can just take another one, write a legacy wrapper and a few patches to maintain compatability with old stuff, deprecate that old stuff, and move on with life.
We all have to take days off, on my days off I don't work, on his days off he's an evil sadistic psychopath.