The ships would decelerate at their destination by releasing a second sail that would reflect the light from the home laser back to the ship
What?!
It launches a second sail ahead of itself, the laser hits that and it reflects the light back; the second sail gets pushed away and lost, but if you can focus it you can keep it pointed at the main craft and slow it down.
I'd draw a bit of ASCII art, but SlashDot is too lame to let me use spaces. Instead, look at something like this paper, describing a roundtrip lightsail.
Don't forget relativity; it's not 30k years for anyone on the ship. Push a decent fraction of c and you can cross just about any distance within your lifetime.
Of course, you'd need a *lot* of energy and some really quite impressive shielding (say, a huge chunk of ice you can refill just about anywhere, ala The Songs of Distant Earth), but given the timeframe we could have to develop that sort of technology, it's not that far out.
Of course, for an external observer, it would be impressive to cross the galaxy in 30k years;)
Depends, do you want to provide any evidence to the contary?
Do I need to read the Bible to realise it's wrong to kill someone* or sleep with someone other than my wife*, not steal*, not commit thoughtcrime*, to love my neighbour and my enemy*, to turn the other cheek*, or that if I ever get the chance, it's a good idea to flood the planet and only tell people I like that they should prepare?
No, I don't think so*. Not that my opinions on the Bible or religion are really up for discussion*:)
Um, who said anything about advice? Frankly I think if we're still around in 50k years, we're not going to be interested in advice from the start of the 21st century.
What they probably *will* be interested in will be the historic value of what thousands of normal people 50k years ago had to say.
As for the Bible, half it's problem is the advice is either useless, counter-productive, out of context, or common sense. Calling it "the most insightful advice possible" is rather laughable IMO.
Anyway, in 50k years, the human race will hopefully have changed a lot. We should have developed a post-scarcity civilization and have a mature social, political and economic system. If not, well, damnit, I wanna know WHY not.
It also supports mixin inheritence, in case you want some more exotic inheritence hierachies; class mixes in Module, and Array mixes in Enumerable, for instance.
I really mean that. Virtually no tools out there produce W3C compliant code? Why?
Simple, your site will be able to displayed less-well on more browsers if you follow what the W3C wants.
So? Who cares if IE3/4 users get the lynx-look? Ok, NN4 is perhaps a little less easy to forget about (luckily for me, it's less than 1% of my userbase), but a gracefully degraded site is still perfectly fine without CSS.
And let's of course not forget that you can use transitional HTML and get a site to display pretty much identically to if you'd used tag soup and STILL be standards compliant.
You're claim that "only geeks care about standards" is an utter joke. Slashdot is a forum dedicated to geeks. Get that? Dedicated to geeks. And yet, does even ONE of their pages conform to the full XHTML 1.0 spec? Or even the HTML 4.01 spec? NO.
Yup, but we all know SlashDot is hardly representative of the cream of the crop in terms of, um, anything.
That's because everyone knows the W3C is a joke.
That's funny. I hang around places with thousands of web developers. I think you'd find your opinion quite rare among them.
They're small, fluffy, cute, clean, affectionate, cheap, intelligent, trainable, and easy to look after.
They only disadantage is they don't live very long (sucks getting attached to one and have it die within 2-3 years), and they will knaw through cables and clothes if you're not careful.
They'll sit on your shoulder while you work, can be trained to come when called, will sit while you stroke them, and come in a wide variety of sizes and colours:)
Yeah, my 350W Enermax PSU comes with a connector that goes to the mobo for the main fan (it has two).
That was, until I blew it up in a dust-removing incident.. bah.
They're great PSU's though. Rated for 100,000 hours, tonnes of gold plated connectors and fan throttle control. Now, if only I could afford to replace it..
All it takes is for someone, somewhere to build an intelligent(ish) self-replicating machine programmed to spread throughout the galaxy - one per interesting star system.
Such a system could cover the entire galaxy in a couple of million years easily (and cheaply after initial design/construction cost). They can do whatever you like; sit and watch, make contact, try to destroy any competitors (The Forge of God style; soon to become a set of movies, yay), and call home (since you end up with a network of them; sure, it'll take a while to get back home, but it's one hell of a cheap way to learn an awful lot about the galaxy).
Given that it only takes one civilization to have done this, and given that our solar system is probably quite interesting given it's layout, I wouldn't put too high odds on there NOT being such a device hanging around near here.
Um, you do realize that you cannot perceive framerates over around 32 fps don't you?
Ignoring for a second that it's quite obvious to anyone who's actually compared 30FPS with 60FPS+ that this isn't true, but if you're averaging 30FPS chances are you're bottoming out on complex scenes around 20 or even 15FPS, and you have zero margin for more complexity.
My old GeForce 256/DDR hit 100FPS in Q3; in MoH:AA using the same engine, there are spots it hits 10FPS. There are even complex Q3 maps which cause it to become noticably jerky, even if average FPS hovers around 100FPS.
Although it's being done with Perl in mind, it's not just the Perl 6 VM; it's actually aimed at pretty much any dynamic language. Hence we should also see backends for Ruby, Python, Basic, any pretty much any language you care to impliment.
There's also talk of Parrot bytecode to Java/CLR bytecode convertors. Interesting stuff, even if we're gonna have to wait ages to actually get something useful.
So it makes a impact crater of about 600 METERS, roughly 2000 feet, in diameter.
That's not a very useful (or accurate) measurement. The actual damage caused is dependent on a lot of things; from which side of the earth it's impacting (forward or away from the rotation?), the angle of approach, the speed of impact, the composition of the impactor (rock? iron? ice?), and if it even manages to HIT the ground.
The Tunguska "impact" didn't actually make it to the ground in one piece; it was an air burst, hence all the flattened trees from the overpreasure -- Just because it didn't make a big impressive crater, doesn't mean you'd want it to detonate anywhere near your house.
2. Why can't Europe get off its butt and save mankind for change? Why is it always the taxpayers of the United States that have to save the Earth?
Well, we keep trying, but every time a film's made about it, it's always the Americans who did it! Frankly, it's getting tiring, so we've decided to just stop telling you lot about every time we deflect an asteroid, destroy an alien race, or send a bunch of people to Mars.
Not to mention that writing for the PS2 is somewhat low-level, so you're likely to be making much more optimized and highly targeted code for it; on a PC you've got DirectX/OpenGL to sit behind, much higher resolutions to deal with, and a much greater diversity of hardware to support.
Plus, a beta is going to be full of debug code, and isn't going to be well optimized.
For some reason, Microsoft's IE/Mac and IE/Win teams are completely different; while IE5/Mac was hailed as having one of the best CSS1 implimentations, IE5/Win was still struggling with the box model (and happily making all your boxes too big, because the IE/Win team can't read, obviously).
So don't go lumping IE/Mac in with IE/Win - they're completely different browsers which happen to share the same name.
mv/usr/obj/usr/obj.old cd/usr/src make update mergemaster -C make buildworld make buildkernel make installkernel make installworld reboot
Although this is just a single user server. mergemaster -C is important with major changes because rc knobs can change occasionally; e.g. sendmail_enable has spawned a lot of friends for all the other daemons it runs.
Perl is quite elegant, and the $ # % all make sense. If you know Perl.
I'm not sure I'd class Perl as being elegant. It's very overcomplex and convoluted to my eye.
Simple things like creating classes, passing large data structures about, gracefully handling error situations, even handling argument lists for subroutines, all require either too much code or too many/\W/'s.
I do admit to seeing some degree of elegance in the whole @foo[] thing, but I'm growing increasingly alergic to touching native data structures directly, so the use of syntax to denote context and type is rapidly loosing it's appeal.
That's not to say I hate Perl, but I do find myself using Ruby for almost all the situations I'd normally use Perl. The resulting code is usually at least as concise and much cleaner. I can't say I miss it, but I don't hate it's very existance like I do VB;)
I imagine the Web Standards site design team had to make a tricky compromise, between the theoretically correct step of sticking to the default browser font size and the more design friendly choice of using "font-size: small".
Am I really the only one to find 1em verdana to be just the right size?
I hate sites that force 11/12px fonts on me - my usual response is to turn off style entirely for these sites if my font-size: 1em !important user.css rule is overridden.
Some design bod on MSDN said people need small text to read comfortably, and that 1/6th of an inch was about optimal; it was ironic that in forcing 12px fonts, the text was actually more like 1/14th of an inch tall, and very tiring to read.
Not exactly. In-between a -RELEASE, the stable branch is not guaranteed to always be buildable or working.
RELENG_4 is the STABLE development branch most people who track STABLE use; this is where prereleases arrive and things are merged from current (MFC); the biggest recent change was an MFC of the new ATA subsystem. New versions of sendmail and smallish changes to the rc system can happen here too.
Although MFC'd stuff is only done so after a lot of testing, and commits to this branch are usually fine, it is still a development branch. Treating it somewhat like Debian/testing is probably a good idea.
For a truely stable up to date system, you should track the RELENG_4_<release> branches, which are the security-update branches for individual releases. Track RELENG_4_6 for 4.6 and you know you won't need to worry too much about running mergemaster to keep/etc in sync, or parts of the base system changing under you in preperation for the next release.
If you track RELENG_4, you should be prepared to at least watch stable@freebsd.org and keep an eye on/usr/src/UPDATING.
And while we're on the subject, remember that cvsup is quite IO intensive; keep your cvsup's conservative. Once a day is usually a bit over the top, and just serves to increase the load on the servers. http://freshports.net/ and ports@freebsd.org are good resources to help decide when it's worth supping.
The Mozilla API model is based on an old and mean-while superseded assumption: that writing software is expensive. In the OpenSource world having to modify a driver because something changed in the kernel, is an advantage not a disadvange, both economically and techically. Proprietary software goes at the tariff of US$ 50-200 per line of debugged code. No such price applies to OpenSource software.
Er, it still costs time and effort to write and debug code, open source or not.
In the mean-time, your constantly changing API's prevent third party code interoperating with your own; you end up with alpha and beta Apache 2 for the rest of eternity, where mod_* only works once in a blue moon when the API versions happen to co-include. You end up with every browser release breaking all your plugins until the maintainers can catch up.
Am I missing something, or am I mistaken in the thought that modern software development has tought us to design well, abstract away details, and decrease coupling?
He seems to be putting a lot of emphasis on embedding Perl inside regexps; while this will make them more powerful and allow you to do more complex things using them, it's increasing coupling between Perl regexp and Perl itself; I don't see the PCRE library and other similar libraries being able to keep up.
Although he mentions he wants regexp to be taken more seriously as a tool for other languages, increasing coupling like this would seem to be a very poor way to go about it.
The new-style regexp's reuse a lot of syntax from the standard regexps; the meaning of many common constructs have changed completely, but I can see it being difficult to actually notice the difference until you've already parsed part of it and gone "Uh?" a few times. And you're not just going to be able to forget about the "old-style" regexps.
Personally I can't help but feel that Larry is the wrong person to make such large changes to the core regexp concepts; they're used heavily in languages other than Perl, but Larry is natuarally doing it in the context of Perl and to hell with any other language. For something so otherwise platform independent, I think this is a bad way to go.
MDI is quite similar (I would say superior) to tabbed browsing. Certainly it doesn't take a great leap to get from MDI to tabs. Either way, Opera's tabs implimentation doesn't break when you open tonnes of them like Mozilla;)
> 1) Good DOM support
Never been interested in DOM support. DHTML is almost universally awful, and none of the sites I use regularly use or depend in it (quite rightly).
> 2) Not crap CSS2 support (Where's IE's and Opera's fixed positioning support?)
Opera does fixed positioning; it just doesn't do overflow, so no emulating frames-based sites in it.
IE supports fixed positioning by switching to default positioning, breaking sites like www.w3.org/Style/; isn't that nice;)
> 3) Image blocking
I have a user css file which blocks most banner adverts:)
> 4) Better cookie management
Never seen the point. I couldn't really care less if someone wants to tag my client and watch what banners I never even see.
> 5) A saner UI. Opera's only good if you really know it.
If you say so. Having to look in Mozilla's dodgy directory hierachy and overwrite one of the files to add my user css file was so much easier than just selecting it in the prefs window in Opera..
> 6) The sidebar (Opera's is nowhere near as customisable)
What, you mean that thing I always turn off in either client?
> 7) The UI takes up less space than Opera
No it doesn't. In fact, with tabs and the document <link> bar it takes about 20px more vertical space as my everything-on Opera display.
> 8) Javascript console > 9) DOM inspector
Not being a DHTML weenie, I can't say I have a use for them.
> 10) XUL
What? How is XUL an advantage? It results in slow, non-standard UI's (Mozilla's URL bar still doesn't work like any other string input bar on Windows. I wonder why).
And don't forget:
11) Exceptional progressive rendering.
On the other hand, Opera has over Mozilla:
Full MDI and SDI tabs, whichever your taste (I can't live without MDI:)
Document/User modes, allowing you to override stupid designers who think 11px fonts on dark backgrounds are the height of readability.
Better printing support (less clipping).
Full screen and projection modes.
Standard OS widgets, not custom, slow, misbehaving ones.
Not an application platform for a load of junk you're never going to use.
More configurable network, cache and security settings, without resorting to user JavaScript files.
Complete page zooming support; not just text size zoom with a tendancy to break layouts.
Doesn't also try to be a complete email/HTML editor/IRC client/tea lady.
It's swings and roundabouts, really; Mozilla and Opera are both good browsers, with different enough approaches to cover most users. If I didn't get a student discount on Opera I'd be using Mozilla.
Slashdot Mirror
Yup.
i x. And memorize it this time ;)
Look at http://foldoc.doc.ic.ac.uk/foldoc/foldoc.cgi?pref
The original Bridge Builder is cool too. Less complex (entirely 2D and only one type of material) but still very fun.
It launches a second sail ahead of itself, the laser hits that and it reflects the light back; the second sail gets pushed away and lost, but if you can focus it you can keep it pointed at the main craft and slow it down.
I'd draw a bit of ASCII art, but SlashDot is too lame to let me use spaces. Instead, look at something like this paper, describing a roundtrip lightsail.
Don't forget relativity; it's not 30k years for anyone on the ship. Push a decent fraction of c and you can cross just about any distance within your lifetime.
;)
Of course, you'd need a *lot* of energy and some really quite impressive shielding (say, a huge chunk of ice you can refill just about anywhere, ala The Songs of Distant Earth), but given the timeframe we could have to develop that sort of technology, it's not that far out.
Of course, for an external observer, it would be impressive to cross the galaxy in 30k years
Depends, do you want to provide any evidence to the contary?
:)
Do I need to read the Bible to realise it's wrong to kill someone* or sleep with someone other than my wife*, not steal*, not commit thoughtcrime*, to love my neighbour and my enemy*, to turn the other cheek*, or that if I ever get the chance, it's a good idea to flood the planet and only tell people I like that they should prepare?
No, I don't think so*. Not that my opinions on the Bible or religion are really up for discussion*
* Various exceptions excluded
Um, who said anything about advice? Frankly I think if we're still around in 50k years, we're not going to be interested in advice from the start of the 21st century.
What they probably *will* be interested in will be the historic value of what thousands of normal people 50k years ago had to say.
As for the Bible, half it's problem is the advice is either useless, counter-productive, out of context, or common sense. Calling it "the most insightful advice possible" is rather laughable IMO.
Anyway, in 50k years, the human race will hopefully have changed a lot. We should have developed a post-scarcity civilization and have a mature social, political and economic system. If not, well, damnit, I wanna know WHY not.
Um, yes it does. For instance:
It also supports mixin inheritence, in case you want some more exotic inheritence hierachies; class mixes in Module, and Array mixes in Enumerable, for instance.
So? Who cares if IE3/4 users get the lynx-look? Ok, NN4 is perhaps a little less easy to forget about (luckily for me, it's less than 1% of my userbase), but a gracefully degraded site is still perfectly fine without CSS.
And let's of course not forget that you can use transitional HTML and get a site to display pretty much identically to if you'd used tag soup and STILL be standards compliant.
Yup, but we all know SlashDot is hardly representative of the cream of the crop in terms of, um, anything.
That's funny. I hang around places with thousands of web developers. I think you'd find your opinion quite rare among them.
They're small, fluffy, cute, clean, affectionate, cheap, intelligent, trainable, and easy to look after.
:)
They only disadantage is they don't live very long (sucks getting attached to one and have it die within 2-3 years), and they will knaw through cables and clothes if you're not careful.
They'll sit on your shoulder while you work, can be trained to come when called, will sit while you stroke them, and come in a wide variety of sizes and colours
Yeah, my 350W Enermax PSU comes with a connector that goes to the mobo for the main fan (it has two).
That was, until I blew it up in a dust-removing incident.. bah.
They're great PSU's though. Rated for 100,000 hours, tonnes of gold plated connectors and fan throttle control. Now, if only I could afford to replace it..
All it takes is for someone, somewhere to build an intelligent(ish) self-replicating machine programmed to spread throughout the galaxy - one per interesting star system.
Such a system could cover the entire galaxy in a couple of million years easily (and cheaply after initial design/construction cost). They can do whatever you like; sit and watch, make contact, try to destroy any competitors (The Forge of God style; soon to become a set of movies, yay), and call home (since you end up with a network of them; sure, it'll take a while to get back home, but it's one hell of a cheap way to learn an awful lot about the galaxy).
Given that it only takes one civilization to have done this, and given that our solar system is probably quite interesting given it's layout, I wouldn't put too high odds on there NOT being such a device hanging around near here.
Ignoring for a second that it's quite obvious to anyone who's actually compared 30FPS with 60FPS+ that this isn't true, but if you're averaging 30FPS chances are you're bottoming out on complex scenes around 20 or even 15FPS, and you have zero margin for more complexity.
My old GeForce 256/DDR hit 100FPS in Q3; in MoH:AA using the same engine, there are spots it hits 10FPS. There are even complex Q3 maps which cause it to become noticably jerky, even if average FPS hovers around 100FPS.
So, no, averaging 32FPS really isn't that great.
[strokes his Ti4200]
Although it's being done with Perl in mind, it's not just the Perl 6 VM; it's actually aimed at pretty much any dynamic language. Hence we should also see backends for Ruby, Python, Basic, any pretty much any language you care to impliment.
There's also talk of Parrot bytecode to Java/CLR bytecode convertors. Interesting stuff, even if we're gonna have to wait ages to actually get something useful.
That's not a very useful (or accurate) measurement. The actual damage caused is dependent on a lot of things; from which side of the earth it's impacting (forward or away from the rotation?), the angle of approach, the speed of impact, the composition of the impactor (rock? iron? ice?), and if it even manages to HIT the ground.
The Tunguska "impact" didn't actually make it to the ground in one piece; it was an air burst, hence all the flattened trees from the overpreasure -- Just because it didn't make a big impressive crater, doesn't mean you'd want it to detonate anywhere near your house.
Well, we keep trying, but every time a film's made about it, it's always the Americans who did it! Frankly, it's getting tiring, so we've decided to just stop telling you lot about every time we deflect an asteroid, destroy an alien race, or send a bunch of people to Mars.
Not to mention that writing for the PS2 is somewhat low-level, so you're likely to be making much more optimized and highly targeted code for it; on a PC you've got DirectX/OpenGL to sit behind, much higher resolutions to deal with, and a much greater diversity of hardware to support.
Plus, a beta is going to be full of debug code, and isn't going to be well optimized.
For some reason, Microsoft's IE/Mac and IE/Win teams are completely different; while IE5/Mac was hailed as having one of the best CSS1 implimentations, IE5/Win was still struggling with the box model (and happily making all your boxes too big, because the IE/Win team can't read, obviously).
So don't go lumping IE/Mac in with IE/Win - they're completely different browsers which happen to share the same name.
A List Apart: Why IE/Mac Matters
Personally I use:
/usr/obj /usr/obj.old /usr/src
mv
cd
make update
mergemaster -C
make buildworld
make buildkernel
make installkernel
make installworld
reboot
Although this is just a single user server. mergemaster -C is important with major changes because rc knobs can change occasionally; e.g. sendmail_enable has spawned a lot of friends for all the other daemons it runs.
I'm not sure I'd class Perl as being elegant. It's very overcomplex and convoluted to my eye.
Simple things like creating classes, passing large data structures about, gracefully handling error situations, even handling argument lists for subroutines, all require either too much code or too many
I do admit to seeing some degree of elegance in the whole @foo[] thing, but I'm growing increasingly alergic to touching native data structures directly, so the use of syntax to denote context and type is rapidly loosing it's appeal.
That's not to say I hate Perl, but I do find myself using Ruby for almost all the situations I'd normally use Perl. The resulting code is usually at least as concise and much cleaner. I can't say I miss it, but I don't hate it's very existance like I do VB
These things are very different; there's a tonne of open source software out there that lots of people use that isn't GPL:
To constantly suggest that the GPL is the One True License is not only wrong, but very damaging, since it undermines perceived choice over licensing.
This is supposed to be a site for vaguely intelligent people; can't we at least make some effort to be more precise in our terminology?
Am I really the only one to find 1em verdana to be just the right size?
I hate sites that force 11/12px fonts on me - my usual response is to turn off style entirely for these sites if my font-size: 1em !important user.css rule is overridden.
Some design bod on MSDN said people need small text to read comfortably, and that 1/6th of an inch was about optimal; it was ironic that in forcing 12px fonts, the text was actually more like 1/14th of an inch tall, and very tiring to read.
Not exactly. In-between a -RELEASE, the stable branch is not guaranteed to always be buildable or working.
/testing is probably a good idea.
/etc in sync, or parts of the base system changing under you in preperation for the next release.
/usr/src/UPDATING.
RELENG_4 is the STABLE development branch most people who track STABLE use; this is where prereleases arrive and things are merged from current (MFC); the biggest recent change was an MFC of the new ATA subsystem. New versions of sendmail and smallish changes to the rc system can happen here too.
Although MFC'd stuff is only done so after a lot of testing, and commits to this branch are usually fine, it is still a development branch. Treating it somewhat like Debian
For a truely stable up to date system, you should track the RELENG_4_<release> branches, which are the security-update branches for individual releases. Track RELENG_4_6 for 4.6 and you know you won't need to worry too much about running mergemaster to keep
If you track RELENG_4, you should be prepared to at least watch stable@freebsd.org and keep an eye on
And while we're on the subject, remember that cvsup is quite IO intensive; keep your cvsup's conservative. Once a day is usually a bit over the top, and just serves to increase the load on the servers. http://freshports.net/ and ports@freebsd.org are good resources to help decide when it's worth supping.
Er, it still costs time and effort to write and debug code, open source or not.
In the mean-time, your constantly changing API's prevent third party code interoperating with your own; you end up with alpha and beta Apache 2 for the rest of eternity, where mod_* only works once in a blue moon when the API versions happen to co-include. You end up with every browser release breaking all your plugins until the maintainers can catch up.
Am I missing something, or am I mistaken in the thought that modern software development has tought us to design well, abstract away details, and decrease coupling?
He seems to be putting a lot of emphasis on embedding Perl inside regexps; while this will make them more powerful and allow you to do more complex things using them, it's increasing coupling between Perl regexp and Perl itself; I don't see the PCRE library and other similar libraries being able to keep up.
Although he mentions he wants regexp to be taken more seriously as a tool for other languages, increasing coupling like this would seem to be a very poor way to go about it.
The new-style regexp's reuse a lot of syntax from the standard regexps; the meaning of many common constructs have changed completely, but I can see it being difficult to actually notice the difference until you've already parsed part of it and gone "Uh?" a few times. And you're not just going to be able to forget about the "old-style" regexps.
Personally I can't help but feel that Larry is the wrong person to make such large changes to the core regexp concepts; they're used heavily in languages other than Perl, but Larry is natuarally doing it in the context of Perl and to hell with any other language. For something so otherwise platform independent, I think this is a bad way to go.
MDI is quite similar (I would say superior) to tabbed browsing. Certainly it doesn't take a great leap to get from MDI to tabs. Either way, Opera's tabs implimentation doesn't break when you open tonnes of them like Mozilla
> 1) Good DOM support
Never been interested in DOM support. DHTML is almost universally awful, and none of the sites I use regularly use or depend in it (quite rightly).
> 2) Not crap CSS2 support (Where's IE's and Opera's fixed positioning support?)
Opera does fixed positioning; it just doesn't do overflow, so no emulating frames-based sites in it.
IE supports fixed positioning by switching to default positioning, breaking sites like www.w3.org/Style/; isn't that nice
> 3) Image blocking
I have a user css file which blocks most banner adverts
> 4) Better cookie management
Never seen the point. I couldn't really care less if someone wants to tag my client and watch what banners I never even see.
> 5) A saner UI. Opera's only good if you really know it.
If you say so. Having to look in Mozilla's dodgy directory hierachy and overwrite one of the files to add my user css file was so much easier than just selecting it in the prefs window in Opera..
> 6) The sidebar (Opera's is nowhere near as customisable)
What, you mean that thing I always turn off in either client?
> 7) The UI takes up less space than Opera
No it doesn't. In fact, with tabs and the document <link> bar it takes about 20px more vertical space as my everything-on Opera display.
> 8) Javascript console
> 9) DOM inspector
Not being a DHTML weenie, I can't say I have a use for them.
> 10) XUL
What? How is XUL an advantage? It results in slow, non-standard UI's (Mozilla's URL bar still doesn't work like any other string input bar on Windows. I wonder why).
And don't forget:
11) Exceptional progressive rendering.
On the other hand, Opera has over Mozilla:
It's swings and roundabouts, really; Mozilla and Opera are both good browsers, with different enough approaches to cover most users. If I didn't get a student discount on Opera I'd be using Mozilla.