The Nasa-Boeing aerospike might have its dimensions measured in inches, but there's lots of them: it translates to about 13 by 7.5 by 7.5 feet.
Turbofans are bulky compared to rockets. Mostly due to the *fan*. Also, most turbofans are designed to run for significant periods (tens or hundreds of hours, at least) between major services, while rockets will only run for tens or hundreds of SECONDS before they need to be stripped and rebuilt. It's like comparing race cars to regular road cars. It all translates to extra size and weight in turbofans.
While Indicated Air Speed is a very useful number aerodynamically (for working out things like stall speed, aerodynamic loading, and so forth), it's not much help when you're trying to work out things like kinetic heating - for that you need to pay attention to the True Air Speed and Mach Number.
Yes, air density is important - you'll get more heating in denser air - but at equivalent IAS values, you'll get more heating at Mach 3 than at Mach 0.5
That said: SSO won't have to deal with kinetic heating anything like that encountered when returning from orbit. At most, I reckon they're just going to need to make some bits of the aircraft out of titanium or steel, rather than aluminium. No big deal.
That is a really warped version of orbital mechanics you're using.
Yes, a lower orbit means you're moving faster, but speeding up will not move you into a lower orbit.
Imagine a ship in a circular orbit. If it fires thrusters to speed up, its orbit will change shape, into an ellipse, with the "lowest" part of the orbit in line with the circlular orbit. As the craft moves from the lowest part of the orbit to the highest, it will be slowed down by gravity, trading kinetic energy for gravitational potential energy. If, at the highest point, the ship fires thrusters to accelerate again, it is possible to end up in a new circular orbit at the higher altitude.
You are correct to state that such a ship will be travelling more slowly in the higher circular orbit than in the lower one, but it got there by using its thrusters to speed up.
The fun involved in doing a rendezvous in orbit is that every time you use thrusters, you change your orbit, and not necessarily in an obvious way.
Speeding up will widen your orbit (taking you "up" or "out") So if your rendezvous target is ahead of you in your orbit, thrusting straight at it will kick you into a wider orbit, and you'll go "over" you target. Slowing down will contract your orbit, taking you under it. Thrusting outwards will widen your orbit, slowing you down; thrusting inwards will contract your orbit, speeding you up.
If your orbit is matched with your target, then every time you thrust, you change orbits, which can take you further away from the target. The way to do it is to get into an orbit which intersects that of your target at the correct moment, and use thrust to match orbits just as the two vehicles intersect.
I have to wonder how it could possibly handle the heat and stresses of atmospheric re-entry.
Re-entry from orbit involves hitting the atmosphere at almost-orbital speeds - about 17,000+ mph.
SSO is designed to fly SUB-orbital. Its re-entry will be MUCH slower. Scaled Composites' website quotes a maximum speed of about 2,500 mph. Kinetic heating shouldn't be a major problem at that sort of speed.
$2M over two years, plus equipment, plus other assistance. That may include developers. Red Hat will continue to contribute developers, has they have done for years. Ditto Sun. Even though they're not mentioned by name, I suspect IBM will too. There will still be developers from the academic world, and unaffiliated individuals pitching in.
>> 1. No Surface to Air missile can reach above 100k feet.
>
> True of shoulder launched missiles, but I'm not so sure about
> things like the Aegis SM2's, or fighter launched air-to-air
> missiles.
No chance. Some of the in-development anti-ballistic missiles, maybe, but there's no operational missile that can even get to 200k feet (that's FORTY MILES UP, folks) let alone chase a target doing 12,500 mph. For reference, that's about 6 times the world air speed record - at nearly double the world altitude record.
Safari weighs in at 7.2 megs, Mozilla is 38.3 megs.
Where in the cosmos did you get those figures from?
The largest version of Mozilla 1.3 Alpha so far is OpenVMS, at 29MB. The Apple version is 18MB, Windows and Linux, 11 and 13mb respectively. Every site I see quoting a figure for Safari says it's 3MB.
As kliklik has already pointed out, it is possible to alter mozilla's useragent string, but if every browser on the web claims to be IE, what incentive is there for anyone to write their websites properly?
That site seem to have been made for IE's poor understanding of CSS rules.:-(
Nope, just dumb webmasters. As far as I can tell, they're using a server-side browser-sniff to send different code to different web browsers. Thing is, they're mis-identifying mozilla (and presumably other mozilla-based browsers) as Netscape. Then, it looks like they still subscribe to the theory that there'e only one Netscape, and are using one tailored for NN4.x, hence it looks crappy.
Opera gets given the IE style sheet, and renders everything just fine. I suspect Moz-based browsers would similarly work fine if they got given that CSS.
If every browser vendor would publish a definite conformance guide like that with their releases and patches, the web would be a better place.
Yes, it would. Opera do. mozilla.org do. If you dig hard enough in MSDN, even Microsoft do.
You've just got to go looking for it all. And whether all those documents are 100% accurate is another question entirely...
The technology exists for laser weapons (basically just high-powered laser pointers), the problem is with a power supply.
Well, power supplies and the Geneva Conventions. Blinding laser weapons (and it's difficult to make a non-blinding one which is still useful as a weapon) are banned. (Protocol IV, 1980)
rogue waves would be easily detected by a terrain following system - same way they detect mountains when flying low over enemy territory, with radar!
Yes, and no. If you're only 20-30 metres up, you'll only be able to see waves from a few miles away with radar - and if you're doing 400-500 mph in a 1500-2000 tonne aircraft, "a few miles" isn't very much warning. Throw in the fact that there are going to be a LOT of waves, of varying height, and it wouldn't be wise to rely on it. You could use radar to get a general idea of the sea condition, and then choose your altitude to give a safety margin above the highest waves.
IIRC a 747 can fly with only one of its four engines working properly.
Well, for certain values of the word "fly", I suppose. It certainly wouldn't be able to engage in any energetic maneuvers, the climb rate would be diabolical (or non-existent) - and if it was one of the outboard engines, steering the beastie would be a little, um, exciting.
Same as any other plane: it slows down a bit. Look at the picture, it's got at least 4 engines (assuming they're not using coax props with multiple engines in each nacelle). If one packs up, it's not a major problem.
There is no margin for error.
Yes there is. 20 feet, at least:). Seriously, when it's cruising, it's not going to be running at full power, so they just throttle the others up a bit, and put up with any loss of speed.
What about rogue waves? It's crusing at 20 feet, and a 40 foot wave pops up?
The document says "as low as 20 feet". If the weather's a bit ropy, they fly a little higher.
For that matter, how calm does the ocean need to be for the ground effect to work?
Not very. What counts is how high your wing is above the "ground". The ground effect starts to be measurable at an altitude of half your wingspan. For it to really work, you want to be somewhat lower. If the ocean is really rough, things will get rather bumpy, though. This thing is projected to have a wingspan of something like 150 metres. That's BIG. It would be getting serious groung effect with its wing 20 metres above the water. That's 65 feet, folks.
Well, maybe you can tell us about what rough edges you are talking about?
Mandrake and SuSE are very polished and EVERYTHING can be done with a GUI.
Weellll, yes, you can do everything through a GUI. But when I tried to get my (short-lived) Mandrake 8 box to talk to my Windows box, Mandrake gave me a pagefull of textboxes labelled with jargon. Dammit, I'm a programmer, a web author and a power user, but I ain't a network engineer. THAT is the kind of rough edge that needs polishing.
Most of my unix experience (I use solaris and occasionally Red Hat in my day job) shows that the biggest problem is that you have to be fairly knowledgable about EVERY PART of the system in order to get things done.
Setting up the most basic network, even through a GUI, means knowing a fair bit about networks. Compare this with Windows XP, where you plug a network cable in, and the OS pops up a dialog saying "hey! a network!", without you even having to touch anything.
Actually Netscape 6.0 (I assume that you mean 6.0 when you say 6) is based on Mozilla M17.
Netscape 6.0 was based on Mozilla M17, plus a bunch of the fixes included in M18, but the mozilla release which most closely matches N6.0 is Moz 0.6. That's more-or-less how moz 0.6 was defined, since it sure didn't fit into any sane release schedule (Source Tree management was a bit wacky in the months between M17 and 0.8)
Netscape 6.1 was based on Moz 0.9.2.
Netscape 6.2 was based on Moz 0.9.4.
Netscape 7.0 preview 1 was based on Moz 1.0rc2.
I've yet to work out exactly what Netscape 7.0 final is based on, but I suspect it's the Mozilla 1.0 branch, in which case N7 will likely be pretty close to the forthcoming Moz 1.0.1.
Create good documentation, written by the least technical person you have to hand who can actually understand how to use the thing. The techies who wrote it always assume too much when they document things - get someone else to write the docs.
Include lots of example code. Write a few samples which use your tool/API/whatever to solve non-trivial problems, and include them, with a detailed step-by-step commentary of what they're up to.
Once you've written all this documentation, give it good indexes, and make it freely available on the web. Particularly, make it very easy indeed to find a good, detailed, description of what your product does, and the examples.
Excel was actually one of the most highly regarded spreadsheets in the 80s
On this, I will stand corrected.
Yes, until WordPerfect released about 4 shitty Windows version in a row.
Was that before or after they had their air supply removed? By 1995, Office was already Market leader. All this happened waaaay before the IE/Netscape thing. WP 5.1 for windows 3.11 worked just fine for me.
MS Office won on its merits.
Office has its plus points (and many of them, paperclip notwithstanding). I do not dispute that it would have gained substantial market share on those merits. But going from nowhere to total domination in 2 years, in a rapidly expanding market, bankrupting most of its previously-market-leading competitors in the process? Honestly, how many products, in any market, are THAT much better than their competition?
And don't take my word for it. Go read the findings of fact. MS Refused to licence Windows 95 to IBM's OEM division because of IBM's support for Lotus, even after Office had "won". (paras 115-132)
Note, not "charged their competitors less for Office", but "refused to sell them windows". In 1995, trying to sell a PC without win95 was like trying to sell a chocolate fireguard.
So, because IBM were supporting Lotus Smartsuite (by this time a relative minnow in the marketplace), MS stuck a knife into IBM's OEM division, and twisted for all it was worth, rather than make loadsamoney selling windows to one of the US's largest OEMs. IBM caved on a few points of disagreement, and still ended up paying more for windows than other OEMs for years afterwards.
[I] do not believe that what was done with IE was anti-competitive in any way.
For example, I think that MS's strongarmed OEM licensing agreements should be regulated or banned altogether.
IE benefited from just such strongarmed licencing practices. Microsoft threatened to withdraw supplies of Windows from OEMs who replaced IE on the desktop with another browser. The whole "Integration with Windows" thing was just a means of justifying it ("want to remove it? Sorry, you can't, it's part of the OS").
Similar threats were made to OEMs who used products competing with MS Office. In the process, Microsoft totally destroyed the market share of Lotus 1-2-3, Quattro Pro, Paradox, Wordperfect and a few others whose names escape me. They very nearly killed Borland as a company, and did kill Wordperfect (who were bought out by Corel). I'm don't recall how Lotus fared (bought out by IBM?).
Those four products, at least, were _all_ respected market leaders. Many people still talk about Wordperfect as some hallowed "golden age" of word processing. Lotus 1-2-3 and Quattro Pro BUILT the spreadsheet market, and Excel was the poor cousin for years, until MS strongarmed it onto half the world's desktops. There is certainly a case for the argument "Netscape would have died anyway" (I used 4.61 - crash central), but not all four of these. In short, MS abused its windows monopoly to totally obliterate its competitors in a very lucrative market.
Microsoft is guilty as hell of abusing its monopoly, and IE vs Netscape is only a sideshow. Even the Findings Of Fact make it clear that Netscape was just "in the way", and MS's real target for their IE behaviour was Java.
There are many who say that the DOJ case was run badly. I think I agree with that in many ways - they went chasing after side issues all the time, and the whole Netscape-Java-Windows-OS-Competitor chain is too tenuous for many people to even follow (never mind the "Java would never have threatened windows anyway" argument). MS Office would have made a much easier target.
you'll get no benefit from `pipelining', which works by downloading several files at a time. It's only useful if you usually have some unused bandwidth.
No. Pipelining involves:
sending multiple HTTP requests in one network packet, and
using a single network connection for multiple files, rather than one file per connection
The net result being that the browser spends far less time messing about with negotiating IP connections or waiting for the server to respond, and more time downloading data.
This has the greatest effect on high latency connections, not low bandwidth ones (though, of course, the two often go hand-in-hand), so that a 28.8 modem to a website hosted by your ISP probably won't show much difference, but a cable modem to a creaking, cruddy server on the other end of the planet will.
Read the
FAQ
Summary:
They can't use Netscape's spellchecker, for legal reasons. Mozilla's own one is still in development and not ready for prime-time yet.
Slashdot Mirror
The Nasa-Boeing aerospike might have its dimensions measured in inches, but there's lots of them: it translates to about 13 by 7.5 by 7.5 feet.
Turbofans are bulky compared to rockets. Mostly due to the *fan*. Also, most turbofans are designed to run for significant periods (tens or hundreds of hours, at least) between major services, while rockets will only run for tens or hundreds of SECONDS before they need to be stripped and rebuilt. It's like comparing race cars to regular road cars. It all translates to extra size and weight in turbofans.
So, what article are you reading?
The 204,420lbf you're quoting is for the Boeing XRS2200 Hydrogen-Oxygen linear aerospike, proposed for the X-33, which never got off the ground.
The little dinky engine powering the rocket mentioned in the article produces 1,000lbf and runs on Ethanol-Oxygen.
While Indicated Air Speed is a very useful number aerodynamically (for working out things like stall speed, aerodynamic loading, and so forth), it's not much help when you're trying to work out things like kinetic heating - for that you need to pay attention to the True Air Speed and Mach Number.
Yes, air density is important - you'll get more heating in denser air - but at equivalent IAS values, you'll get more heating at Mach 3 than at Mach 0.5
That said: SSO won't have to deal with kinetic heating anything like that encountered when returning from orbit. At most, I reckon they're just going to need to make some bits of the aircraft out of titanium or steel, rather than aluminium. No big deal.
That is a really warped version of orbital mechanics you're using.
Yes, a lower orbit means you're moving faster, but speeding up will not move you into a lower orbit.
Imagine a ship in a circular orbit. If it fires thrusters to speed up, its orbit will change shape, into an ellipse, with the "lowest" part of the orbit in line with the circlular orbit. As the craft moves from the lowest part of the orbit to the highest, it will be slowed down by gravity, trading kinetic energy for gravitational potential energy. If, at the highest point, the ship fires thrusters to accelerate again, it is possible to end up in a new circular orbit at the higher altitude.
You are correct to state that such a ship will be travelling more slowly in the higher circular orbit than in the lower one, but it got there by using its thrusters to speed up.
The fun involved in doing a rendezvous in orbit is that every time you use thrusters, you change your orbit, and not necessarily in an obvious way.
Speeding up will widen your orbit (taking you "up" or "out") So if your rendezvous target is ahead of you in your orbit, thrusting straight at it will kick you into a wider orbit, and you'll go "over" you target. Slowing down will contract your orbit, taking you under it. Thrusting outwards will widen your orbit, slowing you down; thrusting inwards will contract your orbit, speeding you up.
If your orbit is matched with your target, then every time you thrust, you change orbits, which can take you further away from the target. The way to do it is to get into an orbit which intersects that of your target at the correct moment, and use thrust to match orbits just as the two vehicles intersect.
Re-entry from orbit involves hitting the atmosphere at almost-orbital speeds - about 17,000+ mph.
SSO is designed to fly SUB-orbital. Its re-entry will be MUCH slower. Scaled Composites' website quotes a maximum speed of about 2,500 mph. Kinetic heating shouldn't be a major problem at that sort of speed.
$2M over two years, plus equipment, plus other assistance. That may include developers. Red Hat will continue to contribute developers, has they have done for years. Ditto Sun. Even though they're not mentioned by name, I suspect IBM will too. There will still be developers from the academic world, and unaffiliated individuals pitching in.
>> 1. No Surface to Air missile can reach above 100k feet. > > True of shoulder launched missiles, but I'm not so sure about > things like the Aegis SM2's, or fighter launched air-to-air > missiles. No chance. Some of the in-development anti-ballistic missiles, maybe, but there's no operational missile that can even get to 200k feet (that's FORTY MILES UP, folks) let alone chase a target doing 12,500 mph. For reference, that's about 6 times the world air speed record - at nearly double the world altitude record.
yes, but by the time they're down to 200,000 feet, they've be going a lot slower.
The largest version of Mozilla 1.3 Alpha so far is OpenVMS, at 29MB. The Apple version is 18MB, Windows and Linux, 11 and 13mb respectively. Every site I see quoting a figure for Safari says it's 3MB.
As kliklik has already pointed out, it is possible to alter mozilla's useragent string, but if every browser on the web claims to be IE, what incentive is there for anyone to write their websites properly?
Mozilla supports Marquee, and
Try this
It's a bit out of date (no details for N7/Opera6/7), but a very detailed breakdown of CSS support in other browsers.
Weellll, yes, you can do everything through a GUI. But when I tried to get my (short-lived) Mandrake 8 box to talk to my Windows box, Mandrake gave me a pagefull of textboxes labelled with jargon. Dammit, I'm a programmer, a web author and a power user, but I ain't a network engineer. THAT is the kind of rough edge that needs polishing.
Most of my unix experience (I use solaris and occasionally Red Hat in my day job) shows that the biggest problem is that you have to be fairly knowledgable about EVERY PART of the system in order to get things done.
Setting up the most basic network, even through a GUI, means knowing a fair bit about networks. Compare this with Windows XP, where you plug a network cable in, and the OS pops up a dialog saying "hey! a network!", without you even having to touch anything.
Netscape 6.1 was based on Moz 0.9.2.
Netscape 6.2 was based on Moz 0.9.4.
Netscape 7.0 preview 1 was based on Moz 1.0rc2.
I've yet to work out exactly what Netscape 7.0 final is based on, but I suspect it's the Mozilla 1.0 branch, in which case N7 will likely be pretty close to the forthcoming Moz 1.0.1.
Create good documentation, written by the least technical person you have to hand who can actually understand how to use the thing. The techies who wrote it always assume too much when they document things - get someone else to write the docs.
Include lots of example code. Write a few samples which use your tool/API/whatever to solve non-trivial problems, and include them, with a detailed step-by-step commentary of what they're up to.
Once you've written all this documentation, give it good indexes, and make it freely available on the web. Particularly, make it very easy indeed to find a good, detailed, description of what your product does, and the examples.
And don't take my word for it. Go read the findings of fact. MS Refused to licence Windows 95 to IBM's OEM division because of IBM's support for Lotus, even after Office had "won". (paras 115-132)
Note, not "charged their competitors less for Office", but "refused to sell them windows". In 1995, trying to sell a PC without win95 was like trying to sell a chocolate fireguard.
So, because IBM were supporting Lotus Smartsuite (by this time a relative minnow in the marketplace), MS stuck a knife into IBM's OEM division, and twisted for all it was worth, rather than make loadsamoney selling windows to one of the US's largest OEMs. IBM caved on a few points of disagreement, and still ended up paying more for windows than other OEMs for years afterwards.
Similar threats were made to OEMs who used products competing with MS Office. In the process, Microsoft totally destroyed the market share of Lotus 1-2-3, Quattro Pro, Paradox, Wordperfect and a few others whose names escape me. They very nearly killed Borland as a company, and did kill Wordperfect (who were bought out by Corel). I'm don't recall how Lotus fared (bought out by IBM?).
Those four products, at least, were _all_ respected market leaders. Many people still talk about Wordperfect as some hallowed "golden age" of word processing. Lotus 1-2-3 and Quattro Pro BUILT the spreadsheet market, and Excel was the poor cousin for years, until MS strongarmed it onto half the world's desktops. There is certainly a case for the argument "Netscape would have died anyway" (I used 4.61 - crash central), but not all four of these. In short, MS abused its windows monopoly to totally obliterate its competitors in a very lucrative market.
Microsoft is guilty as hell of abusing its monopoly, and IE vs Netscape is only a sideshow. Even the Findings Of Fact make it clear that Netscape was just "in the way", and MS's real target for their IE behaviour was Java.
There are many who say that the DOJ case was run badly. I think I agree with that in many ways - they went chasing after side issues all the time, and the whole Netscape-Java-Windows-OS-Competitor chain is too tenuous for many people to even follow (never mind the "Java would never have threatened windows anyway" argument). MS Office would have made a much easier target.
- sending multiple HTTP requests in one network packet, and
- using a single network connection for multiple files, rather than one file per connection
The net result being that the browser spends far less time messing about with negotiating IP connections or waiting for the server to respond, and more time downloading data.This has the greatest effect on high latency connections, not low bandwidth ones (though, of course, the two often go hand-in-hand), so that a 28.8 modem to a website hosted by your ISP probably won't show much difference, but a cable modem to a creaking, cruddy server on the other end of the planet will.
Read the FAQ
Summary:
They can't use Netscape's spellchecker, for legal reasons. Mozilla's own one is still in development and not ready for prime-time yet.