Morgan here from MOTO. Yep, I'm the guy in the video.
Just wanted to chime in and say thanks to everyone for all the good points raised here! Believe it or not, the level of discourse on this site is head and shoulders above the rest of the web in terms of civility, thoughtfulness, and of course, cute snarky humor:)
We're quite aware that this is a very primitive test, and does not fully characterize anything. It's intended as a DIY-means-Do-It-Yourself test for people to get a feel for this stuff in the field, and not much more. Certainly not journal-quality scientific analysis!
There's been a lot of spin applied by the various blog outlets. Don't worry about that nonsense.
We don't mean to puff the iPhone, or to put anyone else down specifically -- we're just trying to demonstrate that there are very real differences in behavior for each of these devices, and that the integration of this behavior is critical in shipping a really high quality product. Full Disclosure: MOTO is a product development consultancy, and we sell our services in this industry. I've been doing this touch stuff for the better part of 10 years. It's definitely in our best interest to show the manufacturers how they can improve, and to try to drive consumers to *demand* better performance.
Anyway, we of course have much more quantitative methods for our actual production work -- there's even an XYZ+pressure digitizer robot behind me in the video -- but this story was about the hands-on approach.
Hope it was at least entertaining/informative!
On a personal note, I have absolutely no allegiance to any of these manufacturers, devices, ecosystems, or otherwise. I'm a luddite rational secular humanist Freedom wonk. Nobody pays us to put these stories up. We can't talk about most of our project work because it's condfidential to our clients -- this is a good way for us to share what we do without exposing anything proprietary, while at the same time driving people to ask the manufacturers for better performance in the future.. We'll put up some more quick tests with other phones as people send them to us. On my list are the Pre, HD2, and a few others. We won't bother with resistive until I've devised a more interesting set of tests to show the advantages and disadvantages of resistive (analog, digital scanned a la Stantum, hybrid a la TouchCo) versus capacitive. Rich stuff.
Next time I might even spend some time to wake the robot up for some testing:)
You'll all be happy to know that the major touch IC vendors have stuff in the pipeline which will provide better multitouch and much better performance in the next year. The silicon in these devices (except the iPhone (I find performance to differ little between 3G and 2G)) is all at least a year behind the state of the art. It's a good year to be an interface nerd.
The "Why" here is all about the read/write speeds of Flash memory... It's just not particularly fast.
Flash memory doesn't "Blit" 512Mbytes over to RAM in a matter of milliseconds. More like a matter of minutes.
And since the HDD already exists, you can just write your RAM to it a la "hibernate" and accomplish the same thing. Plus, your HDD doesn't have wear-levelling issues, and is already part of the PC, and costs orders of magnitude less. Try booting an embedded Linux system that has to pull its rootfs from NAND Flash. Takes for freaking EVER. All the embedded Linux routers out there use NOR for this reason. But NOR parts are slow to write, and aren't even manufactured in densities above about 32 Mbyte last I designed with them.
With HDD read/write speeds at Gb/s, there's just not much advantage. MRAM, if it ever makes it to these scales, would be a real boon -- you could execute straight out of it. NAND Flash is too slow for that.
If you're concerned about the time it takes to write the RAM image to HDD, you could try writing it continually, on-the-fly as you compute, but that seems like a recipe for disaster... eek.
>>A keyboard would have to be absolutely >>revolutionary, fantastic and wonderful
Perhaps something like this?
Now a mere $500, and just as revolutionarily functional as ever.
No joke, I have one for home and one at work -- saved my wrists after everything else failed:)
Na naaaaaaaaaaaa, na na na na na na, na, na naaaaah na nana naaaaaaa... DOOga dooga dooga dooga DOOga dooga dooga dooga! Na NAAaaaaaaaaaa, na na na na na na, na, na naaaaah na nana naaaaaaa... DOOga dooga dooga dooga DOOga dooga dooga dooga!
The consumer can't hack the PureDigital single-use video camera by a longshot.
The guy who got the video off one did it by DESOLDERING the FLASH chip. Most EE's can't do that without destroying the chip, and virtually no one else can. Plus, the camera remains single use -- how many solder/desolder cycles is that chip gonna survive?
The hacks on the old Dakota Digital cameras were great, since all you needed was a cable and some software. I have a few.
But notice that those cameras remain on the market even with the attrition. It was almost in the noise financially speaking.
The video cameras are obviously higher cost-of-goods, and are obviously better-protected. Dollars to donuts even the talented Mr. Maushammer will not successfully crack the video camera, nor even the still-camera-with-display. Meaning "he will not make it a *marketable* hack." Which is all well and good. It's simply too tough to be worth the trouble.
Personally, there's no way I'd bother picking up a camera like this to "hack" it if the hack couldn't provide me with good functionality -- if I have to use a soldering iron to see my "free" videos, I'll pass.
Kudos to Pure Digital for taking the RIGHT economic approach to preventing this kind of arbitrage: Instead of pursuing the obnoxious, expensive avenue of weak protection and litigation (a la DVDJon), they simply make it expensive enough to hack that it's just not worth it.
>The judge in the Napster case defined the litmus >test that has been used by successive generations >of P2P software. His decision is also what got...
^^^ The judge in the Napster case was Marylin Hall Patel. I had lunch with her a couple of times in college when working on a mock trial. Smart lady.
However, Color Kinetics patents are broad enough so as to be construed as covering *simply* PWM-control and turning up Bank A/turning down Bank B. Obviously that would be a very hard sell in court, but it's made an impact on other people's product development decisions.
I can't say who, but a big company I've worked for has specifically backed away from colored LED stuff because of the potential PR nightmare from annihilating the bullshit counterproductive elements in CK's patents.
Fortunately colored LED lighting patents are about as important to the state of the world as P. Diddy's latest car purchase. It's super lame, but IMHO it's not costing us a whole lot.
Personally, I hope all the CK patents get trashed in court.
Moore's "Law" is a Marketing Axiom, not a law of nature or even a good approximation of technical development.
The chip makers have deliberately held their product releases to this rate so that they can continually improve and show growth for Wall Street.
It's a good strategy -- got people to upgrade more often for many years -- only now are they reaching the point where a cheapo home PC has enough horsies to do everything the typical clueless user might with to -- I'm still using 4-year old boxes and doing fine for most everything.
IANASPBMMI, I Am Not A Speech Pathologist But My Mom Is, and for what it's worth, he may simply have been dyslexic and derived the pronunciations from the written forms of the words.
In most of those examples, take note of the way he pulled things apart into component syllables, and essentially just crammed the pronunciation for each syllable together, as each was familiar to him.
One common characteristic of dyslexia is an inability to recognize "whole words" as such -- where you or I see the word "perfunctory", my dyslexic friends see a string of letters with no inherent meaning. Sounding it out is the only way for some people to cope. With hilarious pronunciational (is that a word?) results, admittedly.
I've worked in legal forums on a few occasions (remember Marylin Hall Patel of the Napster ruling?), and the judges/lawyers I've met are insistent on all documents being created/filed in WordPerfect.
Shouldn't say publicly, but it's a small product design consulting firm in San Francisco. We help bridge the gap between good UI design/ID and economically viable mass production in China/Taiwan.
There's really quite a bit of subtlety in translating a high-design product idea into a high-build-quality consumer device that will actually be cheap enough to turn a profit. Takes a lot of experience dealing with ODMs, who in China are VERY good technically, but have little experience making decisions that relate to what Western consumers actually *want*. We help close that gap.
On the other hand, companies that design AND build in Asia are somewhat less beholden to the notion that one must cater to the absolute imbecile, and hence are more likely to implement cool peripheral features that American market-testing would cast aside as too complicated for the target audience. Hence the Japanese cellphones and iRiver music players.
As an example, a number of iRiver players and the iPods are both based the PortalPlayer platform -- start with an iPod and you get the slick out-of-box experience, but drill down into the iRiver firmware and you find a lot of random, cool little features that each might only interest 1% of the user base. Recording, special modes for language-learning materials, sleep-in-N-minutes, etc. Personally, I'm of the opinion that in firmware, more is better as long as it doesn't get in the way of usability. In any case, though, they're based on nearly identical core hardware.
Companies are finally starting to recover from the shock of the iPod's simplicity, and I think we'll start seeing more and more neat useful features and well-designed interfaces as time goes on. That's my main gripe about the iPod versus the Mac -- the Mac has a slick, simple interface with secondary access to good features. The iPod just has the slick simple interface, although the features are starting to show up, generation by generation. Just as it took a few years for cellphones to get to the point where people wanted to download ringtones and wallpaper, it will take a few more for people to figure out other neat ways that their music players can be useful to them.
mod parent up: This is more about players than the Mini, but...
I'm an embedded/firmware developer doing audio players.
The solutions used by virtually *everyone*, Apple included, for their music players are built around overloaded (in a good way) platforms from PortalPlayer and Sigmatel. Typically they're designed so that the platform can do anything from simple flash player to full-fledged multicodec recording and video playback.
The cost of the silicon is basically the same in all of these cases, and makes it possible for the platform developers to support customers in the entire product range (from Shuffle to Portable Media Player) with very little overhead.
The savings come in reducing the peripheral devices that have to go in, as well as the mechanical construction and battery. Flash players, for example, don't have a 20 dollar (in high volume!) LCD, and typically only use a few MB of SDRAM cache since they don't have to worry about HDD duty cycle to conserve power.
I'm not sure of the technical specifics of the hardware, but basically, because we're a tiny (15 people) design firm, most of our office network stuff runs on one or two main servers -- when one server (not the VOIP h/w) went down, anyone calling our front desk got a busy signal.
Obviously our small size allows us to simply suck up the downtime and deal until our IT guy can get it up and running again. Our server failures are generally just HDD crashes and similar MTBF-related hardware issues. Not overloading, that's for sure.
However, there's more to it -- certainly the capability exists for reliable VOIP hardware -- as an EE, I'm confident of that. Our problem is that as early adopters we're subject to all of the buggy hardware/firmware (including the phones!) that Avaya is effectively beta-testing on us.
I *have* been satisfied with call quality, and haven't had any trouble with dropped calls. The friends of mine with residential VOIP are *crazy* about it.
the citations in the dissenting opinion. To prove his point that P2P filesharing is damaging record sales, Murphy cites two NEWSPAPER ARTICLES,
one from the NYT and one from the Wash. Post. He doesn't cite any economic study or any actual data beyond speculative news articles.
I'd like to see them get their just desserts too, but:
If he's selling it as Vioxx and shipping something else, that's one thing. But if he's not, I call it Natural Selection... People need to take responsibility for themselves. If you knowingly self-medicate and you die, too bad -- should have consulted your physician. If someone misrepresents, *then* there's a case for liability.
This is a question of economics -- absurd price inequities in pharma have created a black market between the US and just about everywhere else. It may not be legal, but it's hardly new or surprising -- it's just the market setting the REAL price. Where's that globalization they promised me?;) What a mess...
Parachute performance remains, as always, NOTHING like airplane performance -- even the slickest, fasted skydiving rigs don't glide better than a few-to-one ratio. Top paragliders, which fly slowly at 15-35mph, max out at 9-to-1 glide ratio.
That aside: Landing an unpowered aircraft is all about flare timing and landing gear -- you've got to dissipate as much kinetic and potential energy as possible before you have to interface with the ground. Especially when you don't have wheels! If you've got wheels and a smooth surface, you can land a five-hundred ton plane with high reliability. Without wheels, and on a really rough surface (say a mountainside), only the best hang glider pilots (wing loading ~1lb/sqft, total system mass ~220lbs) can consistently land safely, and even then it's tight. Ever notice how swooping contests tend to happen near lakes?:)
In hang gliders and paragliders both, and increase in ground speed is invariably met with a decrease in the size of the "flare window" -- i.e. the interval in which you can initiate the flare and a) not gain too much altitude before coming back down b) not be too late and nose in due to stall.
It's a very delicate dance. Given that skydivers swooping routinely land at 40+ mph, I don't think it's impossible, but the amount of control you have in a wingsuit is pretty minimal. I predict grievous injury, but probably not death.
Landing on water, however, seems *entirely* plausible.
Slashdot Mirror
^^ EXACTLY. This AC knows what's up.
Hi Everyone,
Morgan here from MOTO. Yep, I'm the guy in the video.
Just wanted to chime in and say thanks to everyone for all the good points raised here! :)
Believe it or not, the level of discourse on this site is head and shoulders above the rest of the web in terms of civility, thoughtfulness, and of course, cute snarky humor
We're quite aware that this is a very primitive test, and does not fully characterize anything.
It's intended as a DIY-means-Do-It-Yourself test for people to get a feel for this stuff in the field, and not much more.
Certainly not journal-quality scientific analysis!
There's been a lot of spin applied by the various blog outlets. Don't worry about that nonsense.
We don't mean to puff the iPhone, or to put anyone else down specifically -- we're just trying to demonstrate that there are very real differences in behavior for each of these devices, and that the integration of this behavior is critical in shipping a really high quality product.
Full Disclosure: MOTO is a product development consultancy, and we sell our services in this industry. I've been doing this touch stuff for the better part of 10 years.
It's definitely in our best interest to show the manufacturers how they can improve, and to try to drive consumers to *demand* better performance.
Anyway, we of course have much more quantitative methods for our actual production work -- there's even an XYZ+pressure digitizer robot behind me in the video -- but this story was about the hands-on approach.
Hope it was at least entertaining/informative!
On a personal note, I have absolutely no allegiance to any of these manufacturers, devices, ecosystems, or otherwise. I'm a luddite rational secular humanist Freedom wonk.
Nobody pays us to put these stories up. We can't talk about most of our project work because it's condfidential to our clients -- this is a good way for us to share what we do without exposing anything proprietary, while at the same time driving people to ask the manufacturers for better performance in the future..
We'll put up some more quick tests with other phones as people send them to us.
On my list are the Pre, HD2, and a few others. We won't bother with resistive until I've devised a more interesting set of tests to show the advantages and disadvantages of resistive (analog, digital scanned a la Stantum, hybrid a la TouchCo) versus capacitive. Rich stuff.
Next time I might even spend some time to wake the robot up for some testing :)
You'll all be happy to know that the major touch IC vendors have stuff in the pipeline which will provide better multitouch and much better performance in the next year.
The silicon in these devices (except the iPhone (I find performance to differ little between 3G and 2G)) is all at least a year behind the state of the art.
It's a good year to be an interface nerd.
Thanks for all the feedback!
morgan
The "Why" here is all about the read/write speeds of Flash memory... It's just not particularly fast.
Flash memory doesn't "Blit" 512Mbytes over to RAM in a matter of milliseconds. More like a matter of minutes.
And since the HDD already exists, you can just write your RAM to it a la "hibernate" and accomplish the same thing. Plus, your HDD doesn't have wear-levelling issues, and is already part of the PC, and costs orders of magnitude less. Try booting an embedded Linux system that has to pull its rootfs from NAND Flash. Takes for freaking EVER. All the embedded Linux routers out there use NOR for this reason. But NOR parts are slow to write, and aren't even manufactured in densities above about 32 Mbyte last I designed with them.
With HDD read/write speeds at Gb/s, there's just not much advantage. MRAM, if it ever makes it to these scales, would be a real boon -- you could execute straight out of it. NAND Flash is too slow for that.
If you're concerned about the time it takes to write the RAM image to HDD, you could try writing it continually, on-the-fly as you compute, but that seems like a recipe for disaster... eek.
y'know, if you just click the "I've already donated" button, it'll stop bugging you ;)
>>A keyboard would have to be absolutely
:)
>>revolutionary, fantastic and wonderful
Perhaps something like this?
Now a mere $500, and just as revolutionarily functional as ever.
No joke, I have one for home and one at work -- saved my wrists after everything else failed
Don't forget the bass beat!
Na naaaaaaaaaaaa, na na na na na na, na, na naaaaah na nana naaaaaaa...
DOOga dooga dooga dooga DOOga dooga dooga dooga!
Na NAAaaaaaaaaaa, na na na na na na, na, na naaaaah na nana naaaaaaa...
DOOga dooga dooga dooga DOOga dooga dooga dooga!
would include a soda-can gripper as primary equipment on an $800 swarmbot...
A good point.
Worth noting:
The consumer can't hack the PureDigital single-use video camera by a longshot.
The guy who got the video off one did it by DESOLDERING the FLASH chip. Most EE's can't do that without destroying the chip, and virtually no one else can. Plus, the camera remains single use -- how many solder/desolder cycles is that chip gonna survive?
The hacks on the old Dakota Digital cameras were great, since all you needed was a cable and some software. I have a few.
But notice that those cameras remain on the market even with the attrition. It was almost in the noise financially speaking.
The video cameras are obviously higher cost-of-goods, and are obviously better-protected. Dollars to donuts even the talented Mr. Maushammer will not successfully crack the video camera, nor even the still-camera-with-display. Meaning "he will not make it a *marketable* hack." Which is all well and good. It's simply too tough to be worth the trouble.
Personally, there's no way I'd bother picking up a camera like this to "hack" it if the hack couldn't provide me with good functionality -- if I have to use a soldering iron to see my "free" videos, I'll pass.
Kudos to Pure Digital for taking the RIGHT economic approach to preventing this kind of arbitrage:
Instead of pursuing the obnoxious, expensive avenue of weak protection and litigation (a la DVDJon), they simply make it expensive enough to hack that it's just not worth it.
>The judge in the Napster case defined the litmus
;)
>test that has been used by successive generations
>of P2P software. His decision is also what got...
^^^
The judge in the Napster case was Marylin Hall Patel. I had lunch with her a couple of times in college when working on a mock trial. Smart lady.
Lady being the operative term here
However, Color Kinetics patents are broad enough so as to be construed as covering *simply* PWM-control and turning up Bank A/turning down Bank B. Obviously that would be a very hard sell in court, but it's made an impact on other people's product development decisions.
I can't say who, but a big company I've worked for has specifically backed away from colored LED stuff because of the potential PR nightmare from annihilating the bullshit counterproductive elements in CK's patents.
Fortunately colored LED lighting patents are about as important to the state of the world as P. Diddy's latest car purchase. It's super lame, but IMHO it's not costing us a whole lot.
Personally, I hope all the CK patents get trashed in court.
Moore's "Law" is a Marketing Axiom, not a law of nature or even a good approximation of technical development.
The chip makers have deliberately held their product releases to this rate so that they can continually improve and show growth for Wall Street.
It's a good strategy -- got people to upgrade more often for many years -- only now are they reaching the point where a cheapo home PC has enough horsies to do everything the typical clueless user might with to -- I'm still using 4-year old boxes and doing fine for most everything.
is one of the 600,426,974,379,824,381,952 ways to spell \/14grA
dictionary-attack that, G-man!
IANASPBMMI,
I Am Not A Speech Pathologist But My Mom Is, and
for what it's worth, he may simply have been dyslexic and derived the pronunciations from the written forms of the words.
In most of those examples, take note of the way he pulled things apart into component syllables, and essentially just crammed the pronunciation for each syllable together, as each was familiar to him.
One common characteristic of dyslexia is an inability to recognize "whole words" as such -- where you or I see the word "perfunctory", my dyslexic friends see a string of letters with no inherent meaning. Sounding it out is the only way for some people to cope. With hilarious pronunciational (is that a word?) results, admittedly.
people offering free expensive things are usually lying!
the standard in legal documents for many years.
I've worked in legal forums on a few occasions (remember Marylin Hall Patel of the Napster ruling?), and the judges/lawyers I've met are insistent on all documents being created/filed in WordPerfect.
Shouldn't say publicly, but it's a small product design consulting firm in San Francisco. We help bridge the gap between good UI design/ID and economically viable mass production in China/Taiwan.
There's really quite a bit of subtlety in translating a high-design product idea into a high-build-quality consumer device that will actually be cheap enough to turn a profit. Takes a lot of experience dealing with ODMs, who in China are VERY good technically, but have little experience making decisions that relate to what Western consumers actually *want*. We help close that gap.
On the other hand, companies that design AND build in Asia are somewhat less beholden to the notion that one must cater to the absolute imbecile, and hence are more likely to implement cool peripheral features that American market-testing would cast aside as too complicated for the target audience. Hence the Japanese cellphones and iRiver music players.
As an example, a number of iRiver players and the iPods are both based the PortalPlayer platform -- start with an iPod and you get the slick out-of-box experience, but drill down into the iRiver firmware and you find a lot of random, cool little features that each might only interest 1% of the user base. Recording, special modes for language-learning materials, sleep-in-N-minutes, etc. Personally, I'm of the opinion that in firmware, more is better as long as it doesn't get in the way of usability. In any case, though, they're based on nearly identical core hardware.
Companies are finally starting to recover from the shock of the iPod's simplicity, and I think we'll start seeing more and more neat useful features and well-designed interfaces as time goes on. That's my main gripe about the iPod versus the Mac -- the Mac has a slick, simple interface with secondary access to good features. The iPod just has the slick simple interface, although the features are starting to show up, generation by generation. Just as it took a few years for cellphones to get to the point where people wanted to download ringtones and wallpaper, it will take a few more for people to figure out other neat ways that their music players can be useful to them.
mod parent up:
This is more about players than the Mini, but...
I'm an embedded/firmware developer doing audio players.
The solutions used by virtually *everyone*, Apple included, for their music players are built around overloaded (in a good way) platforms from PortalPlayer and Sigmatel. Typically they're designed so that the platform can do anything from simple flash player to full-fledged multicodec recording and video playback.
The cost of the silicon is basically the same in all of these cases, and makes it possible for the platform developers to support customers in the entire product range (from Shuffle to Portable Media Player) with very little overhead.
The savings come in reducing the peripheral devices that have to go in, as well as the mechanical construction and battery. Flash players, for example, don't have a 20 dollar (in high volume!) LCD, and typically only use a few MB of SDRAM cache since they don't have to worry about HDD duty cycle to conserve power.
Don't get out to California much, eh? ;)
There hasn't been a tiny, noisy 1BR condo for sale in Silicon Valley for less than about $350k for at least 5 years. Even then it's slim pickins.
Sigh.
Me, I'd take rural oregon any day if I could just move my job up there. Real purty.
>> would probably prefer to not risk their jobs."
/. editors not to split their infinitives...
would probably prefer the
Most definitely. We're subject to all of Avaya's beta-ass h/w, f/w and s/w. It's a half-baked $HARDNESS/w-fest.
I'm not sure of the technical specifics of the hardware, but basically, because we're a tiny (15 people) design firm, most of our office network stuff runs on one or two main servers -- when one server (not the VOIP h/w) went down, anyone calling our front desk got a busy signal.
Obviously our small size allows us to simply suck up the downtime and deal until our IT guy can get it up and running again. Our server failures are generally just HDD crashes and similar MTBF-related hardware issues. Not overloading, that's for sure.
However, there's more to it -- certainly the capability exists for reliable VOIP hardware -- as an EE, I'm confident of that. Our problem is that as early adopters we're subject to all of the buggy hardware/firmware (including the phones!) that Avaya is effectively beta-testing on us.
I *have* been satisfied with call quality, and haven't had any trouble with dropped calls. The friends of mine with residential VOIP are *crazy* about it.
My company switched to a VOIP solution a few months ago.
Now, when our server crashes, our phones go out.
Sweet.
the citations in the dissenting opinion.
To prove his point that P2P filesharing is damaging record sales, Murphy cites two NEWSPAPER ARTICLES,
one from the NYT and one from the Wash. Post. He doesn't cite any economic study or any actual data beyond speculative news articles.
Not a very impressive rationalization.
I'd like to see them get their just desserts too, but:
;) What a mess...
If he's selling it as Vioxx and shipping something else, that's one thing. But if he's not, I call it Natural Selection... People need to take responsibility for themselves. If you knowingly self-medicate and you die, too bad -- should have consulted your physician. If someone misrepresents, *then* there's a case for liability.
This is a question of economics -- absurd price inequities in pharma have created a black market between the US and just about everywhere else. It may not be legal, but it's hardly new or surprising -- it's just the market setting the REAL price. Where's that globalization they promised me?
IAAHG/PGP (I am a hang glider/paraglider pilot).
:)
Parachute performance remains, as always, NOTHING like airplane performance -- even the slickest, fasted skydiving rigs don't glide better than a few-to-one ratio. Top paragliders, which fly slowly at 15-35mph, max out at 9-to-1 glide ratio.
That aside:
Landing an unpowered aircraft is all about flare timing and landing gear -- you've got to dissipate as much kinetic and potential energy as possible before you have to interface with the ground. Especially when you don't have wheels! If you've got wheels and a smooth surface, you can land a five-hundred ton plane with high reliability. Without wheels, and on a really rough surface (say a mountainside), only the best hang glider pilots (wing loading ~1lb/sqft, total system mass ~220lbs) can consistently land safely, and even then it's tight. Ever notice how swooping contests tend to happen near lakes?
In hang gliders and paragliders both, and increase in ground speed is invariably met with a decrease in the size of the "flare window" -- i.e. the interval in which you can initiate the flare and
a) not gain too much altitude before coming back down
b) not be too late and nose in due to stall.
It's a very delicate dance. Given that skydivers swooping routinely land at 40+ mph, I don't think it's impossible, but the amount of control you have in a wingsuit is pretty minimal. I predict grievous injury, but probably not death.
Landing on water, however, seems *entirely* plausible.
"Good article over at CNet regarding Microsoft's infernal IT practices. Some intriguing "
nothing to see here...