> The X-Windowing-System has come with xmag virtually for ever.
xmag is also quite inferior to the magnifier thingy that comes with Windows 98.
However, X also has the ability to set a desktop any size you like and then zoom in to view part of it at 640x480 with Ctrl-Alt-+, which with a large monitor should provide quite good magnification. This is much closer to the functionality that the Windows magnifier provides, because you can easily pan around. If it were possible to zoom in to even lower resolutions (320x200 comes to mind) it would be better. (Maybe it _is_ possible to do that, by writing a custom modeline for those resolutions, but the GUI config tools for X that ship with Mandrake don't make it easy to set that up.)
Really, xmag ought to be improved (or replaced) to be resizeable and support a follow-the-mouse-cursor mode.
Unfortunately, it doesn't seem to work with the regular cursor movement keys, only with the keypad, but I find it necessary to remap the keypad to emulate the regular cursor keys, on the grounds that otherwise most X apps don't support the keypad properly for cursor movement (which is *way* more important to me than mouse movement). (Just using the non-pad cursor keys is unacceptable because they're arranged bady.)
I suppose I could map things so that the regular arrows emulate the pad arrows... I never had a reason before (having no use for the non-cursor functionality of the pad), but this might be a good enough reason.
Do the pad home/end/pgup/pgdn keys do anything meaningful with the shift+numlock keyboard-mouse mode turned on?
> there -will- be a future for anyone who fell in love with BeOS
I believe the most meaningful future contribution of BeOS is the set of ideas it has made available to the developer community. The BeOS was *innovative*, innovative in ways no current major OS is innovative.
We're already seeing ideas from the BeOS incorporated into other systems. Filesystem journaling is a good example. The new (yet to be released) MS filesystem is another, perhaps better, example. Have you noticed people in the OSS community talking about threading lately, and about making things thread-safe? Heck, have you seen processors for sale with hyperthreading technology? Thank Be for demonstrating the performance advantages of a heavily threaded system. These are not pie in the sky but real, concrete benefits that we have because of the BeOS.
There's another thing current OSes need to learn from BeOS; Linux has started, but there's a long way to go: better on-the-fly hardware recognition. It's nothing to take a hard drive with the BeOS on it out of one computer and stick it in another computer, turn it on, and just start using it. No fooling with driver CDs, no reconfiguring X, no "New Hardware Found" dialog boxes, just turn it on and use it. (This is also part of what makes BeOS so easy to install.) Kudzu and HardDrake are starting to be pretty impressive, but the BeOS still has them beat on this point, so there's room to grow.
Okay, so it's missing some things like a working security model, the ability to set global color preferences, and (since it's not 1998 anymore) support for modern hardware. But we can learn from the things it's *not* missing.
I still want a versioned filesystem. ITS had it in the days of yore; VMS has had it for a good while too. C'mon, somebody get on the stick and put it in a modern OS.
I think Microsoft has more than nine lawyers. If they're more formidable than ordinary orcs, perhaps they're Uruk Hai or trolls or something. The Nazgul metaphor must be reserved for something there are only a handful of, like totally satisfied Microsoft customers or rock-solid secure Windows servers. Say, where's Eowyn?
> but we should really be debating how we get this right on an > OSS platform
For starters we get ourselves a VHLL or three and stop writing everything in %$@! C and C++. There will still be bugs, of course. But I'm getting *really* tired of hearing about newly discovered buffer overruns; that Can't Happen to an application written in any decently modern language.
Re:Shower or bathe at least once a week - WTF ?!
on
Cubicle Etiquette?
·
· Score: 1
> But that is precisely why extensions exist. So that you don't > have to have all of those features installed.
The problem is not with the fact that the extensions are optional. The problem is that the install process for them is far too cumbersome for people who want most of them. I had to spend over half an hour installing them at home (dialup), and half that time at work. That's time I had to be actively involved with the install process, because there were regular dialogs to answer new install links to click. It ought to be possible to check some checkboxes for which ones to install, click a couple of buttons and then *do something else* while it happens.
> Personally, I think that's how it should stay. Keep the browser > lean and let people modify it to install whichever extensions > they want.
Fine, but I should be able to install a bunch of them at one go. This nonsense about spending fourty-five minutes installing thirty individual extensions one-by-one, answering 2-3 dialog boxes for *each* of them (and needing to read these boxes, since the question about which directory to install into has the yes/no backwards for about half of them) is for the birds (harhar).
I want one button I can click to install them all, and one dialog to answer for *all* of them whether to install in my profile or globally. Then if I want I can disable two or three of the misguided ones (like the one that turns the alt attribute for img elements into an abbrev tag like in Netscape 4).
People who only want one or two extensions should still be able to install them individually, of course.
> Perl 4 had Unicode support the day it was ported to plan9
Who on earth needed Unicode support in the days of Perl4? Unicode didn't gain buzzword status until a couple of years ago, right about the same time as XML. I'm still not sure *why* we need it, other than that of course everything has to support it these days to be considered modern and hip. Bah. XML at least is occasionally useful.
Regarding screen sizes: the same people who complain about small
type will insist they need a small screen, that 18" is too large.
They cannot coherently explain why smaller things are easier to
see, but a large percentage of them insist on it, including almost
everyone who wears bifocals. Many of them use 14" viewables and
don't totally maximize the browser window.
In order to compensate for the small size of their screens, they
will cut the resolution to 640x480, and I'm convinced they'd set it
smaller than that if Windows would let them. Then on top of that
they will only effectively use about half of the screen, saying
that they can't see the top part, the bottom part, or whatever.
The implication is that if this is your target market, your site
needs to be usable at about 600x200. Whee. If you plan it right
from the start, it is entirely possible to design a site that looks
decent at that size and still scales and looks okay at much higher
resolutions, but with bitmap-type graphics there are limits; it's
going to look quite stretched at 1600x1200.
There are two ways around this: one is to make your graphics scale,
and the other is to build your site mostly out of text, maybe using
graphics for borders and backgrounds and stuff.
Making graphics scale *properly* means vector graphics. When you
run across a vector graphics format with wide browser support please
let me know, as I'd be very interested. As a kludge, you can use
bitmapped graphics (PNG or whatever) and assign relative widths.
I have done this once or twice (width="100%" in my case)
in a pinch,
but if you try it you will immediately see the problem. Depending
on the graphic it might be okay for some things, but it's definitely
not a general solution to the whole issue.
At this time, my recommendation for sites that need to scale well
to different resolutions is to make heavy use of text and style
sheets and use a few strategically-placed graphics to spruce things
up without interfering too much with scalability. For example, a
background graphic that can be tiled will accomodate different
resolutions fairly well. Narrow borders that scale or repeat in
one direction are another fine example. A medium-sized logo that
can be centered at the top, above the rest of the content, may be
designed such that it looks fine surrounded by varying amounts of
whitespace. And so on.
As far as text, use the relative size attributes to make some text
larger than the rest as necessary, but don't fix hard sizes, as
some legacy browsers[1] then won't let the user scale the text;
with relative sizes the browser will pick up and use the user's
base font size.
Some of your layout problems can be lessened by use of alpha-channel
transparency. This doesn't work with all browsers, though; it works
with all browsers based on Mozilla.org code, recent versions of
Opera, and possibly certain others, but not for example with old
versions of Netscape. There is a kludge to make it work with some
versions of MSIE, but this fails sometimes
depending on the user's settings and in any case will not
work with old versions of MSIE. If you are interested in pursuing
alpha channel stuff, I have some examples up here:
http://cgi.galion.lib.oh.us/test/
Especially see
GPL-plus.html (for an example how how it helps layouts) and png-alpha/png-alpha-demo-hacked.html (for a demo of the transparency channel itself). There is also a
really cool demo here,
but that one has not been hacked to work with MSIE.
One final piece of advice: test your site at at least three
resolutions, including 640x480 and 1280x1024.
> All I know is I had a giant plug with four prongs on it.
Oh, those. Incidentally, you can make a standard phone line work with only two of those four wires. (This is still true with an RJ12 connector; you only need two of the wires, for a voice line.)
If you think it's bad trying to keep RJ12 and RJ45 straight, you ought to have to deal with the *other* kinds of modular connectors. RJ12 has four wires and RJ45 has eight, but did you know, there are two different kinds with six wires, differing only by the placement of the little clip thingy that holds them in the socket? The one with the centered clip is RJsomething (I forget the number, but it's between 12 and 45); the off-center one is called MMJ or DEC423. I have a crosspinned inline coupler for this type... and a real, non-historical use for it.
> any sentence challenging English usage or pronunciation that ends > in a preposition needs revisiting
"at" in that sentence is not functioning as a preposition; it is functioning as the complementary part of the verb. Besides, the rule "never end a sentence with a preposition" is significantly oversimplistic; the correct rule is that the words in a prepositional phrase must be kept together, in this order: the preposition first, followed by any standard attributive adjectives modifying the object, followed by the object itself, followed by any additional modifiers (such as modifying phrases or clauses). The occurrance of other words, not part of any prepositional phrase, that in other circumstances might be used as prepositions, is irrelevant.
> You seem to be forgetting that a critical part of the > laminating process is extreme heat.
The big, fast, convenient laminators work that way. The cheapo ones just use two rolls of clear contact paper (one top and one bottom), rollers, and a hand crank. Should be fine.
Basically, common sense...
on
Cubicle Etiquette?
·
· Score: 2, Insightful
* Think before you act. If it would annoy *you*, chances are
it may annoy the guy nextdoor, also. * Keep the noise down. If you must have sound from your PC or
stereo, get headphones. If you need to carry on a conversation,
go to the person, rather than yelling across the room. * Don't do anything you see done in a Dilbert cartoon. * Shower or bathe at least once a week whether you need it or not.
There may be a handful of other things peculiar to the environment, but I'm certain that you can get 95% of the way there with basic everyday common sense.
This is an excellent opportunity for web-standards zealots to roll out the FUD. "If your website uses plugins, you could be sued for patent infringement, forced to pay back royalties, and put out of business! Quickly, convert all your websites to XHTML/CSS and be safe from the patent lawyers!"
IE loads significant parts of itself at system start time, so it has less to load when you launch it. Mozilla (SeaMonkey) used to have an equivalent feature, but circa 1.0 it was tworked (to improve stability; apparently there were a number of memory-handling bugs that could be worked around by forcing a reload of everything) and hasn't really worked properly since. Firebird has AFAIK never had this feature; I don't know whether there are any plans to add it.
> I have not tried Mozilla on a sub Ghz machine, nor do I intend to. > Some of us like to keep current with the times... my THREE previous > machines were faster than 1ghz.
I recently (this past spring) upgraded. Previously I was using a PentiumII/233, and now it's a Pentium4/2.somethingGHz with the 800 MHz FSB. I've noticed very significant performance improvements for some applications -- ifile, for example -- and for games. Mozilla, however, performs pretty much exactly the same. This is probably because I had my PII/233 maxed out on 512MB of RAM, which was enough to make Mozilla happy. Browsing the web uses a lot more RAM than CPU. Also, my internet connection hasn't changed, nor has my hard drive changed. My RAM is now DDR, supposedly faster than the SDRAM I had before, but that doesn't seem to make as much difference.
Now, a while back when I added RAM to my PII/233 to bring it from 192 to 512MB, I did notice major performance improvements for Mozilla at that time. I conclude that RAM is a major factor for Mozilla performance, and CPU isn't. There are other major factors besides RAM; your internet connection obviously would be one; CPU speed doesn't seem to be, though.
The difference is, FireBird has a smaller memory footprint than SeaMonkey. As such, if you have a marginal amount of RAM, FireBird may fit without swapping where SeaMonkey doesn't. This will result in a significant perceived performance improvement. If you have plenty of RAM, however, SeaMonkey will use more of it (and maybe take a bit longer to start the application initially), but apart from that the rendering speed will be the same -- they are both using Gecko, after all.
> Also, since I keep mozilla open all the time it isn't too > important how long it takes to load.
Indeed. I haven't restarted my browser since the big power outage, and that's pretty typical for me. Often I'll leave it running from the day I install it until the day I install the next version. It's too much of a pain to simultaneously bring everything I was doing to a close all at once. The ability to bookmark tabsets helps somewhat with that, but you still lose your place, any sessions, forms you were filling out, and so on. I'm tempted to skip 1.5 and wait for the next version, just because I don't feel like restarting my browser again yet...
> The sad, sad news is that Firebird and Thunderbird will not made > it into 1.5:-(
If you've been testing Firebird and Thunderbird this is good news. They're not ready. Firebird is getting there, and hopefully will be ready to replace Navigator by 1.6 time, but SeaMonkey really can't be put out to pasture if only Navigator has been adequately replaced. Thunderbird... well, it still needs a lot of work. Also, Sunbird needs to be working before SeaMonkey can be dropped.
Actually, Firebird has most of the features Navigator has, *if* you install a metric tonne of Extensions. (This is a major issue, however; it takes considerably longer and *many* times more clicking to download and install all those extensions as compared to just downloading and installing the entire SeaMonkey suite. A solution needs to be worked out wherein many extensions can be downloaded and installed in one go.) Even with all of the extensions, though, FB is still missing a couple of very major features, like the DOM inspector (which is dogfood, or should be -- it's painful to do any work on themes without it; it's quite handy for web development also).
> Once Gimp is updated to use the newer version of GTK, I'll > be very happy though:o)
The new version of Gimp (1.3) uses the new GTK. You will, however, want to keep your existing Gimp 1.2 install around also. The new Gimp has some cool features, but at this point it's still also missing some things that are present in 1.2.
> Remember "social engineering" only works on people with social > skills! We read BOFH articles in the same way as "HOW-TO" documents!
User: I'm having a little trouble starting up Notepad... BOFH: That's because we're standardising everyone on two text
editors, to maintain consistency across the network. We
upgraded the Windows systems from Notepad to EDLIN last
night during overnight processing. User: But I don't know how to use EDLIN! BOFH: Whose fault is that? User: You said two text editors. What's the other one? BOFH: The Unix systems all have sed. You wouldn't believe
the whining we got from the vim nerds and Emacs geeks,
but they'll get over it.
> at this point i would venture to say that most average people > would be lost without OE and O when it comes to email.
Bunk. I install Pegasus Mail for end users -- users who don't know how to copy and paste and are afraid to learn, people who don't understand multitasking, don't know what the taskbar is for, believe they need to "x out" of one app to get to something else -- and they don't have nearly as much trouble with Pegasus mail as the people have who are using Outlook -- and I'm not talking about trouble with viruses; I'm talking about people being able to figure out how to read their mail. Pegasus has been that easy to use since before there was Outlook. Yes, it has advanced features, but you have to go digging through the menus and the preferences dialogs to find them.
Now, some really powerful mailreaders aren't appropriate for end users. The one I use (Gnus) is certainly not right for them. But mailreaders exist that are easier to use than Outlook and better. And free, unless you think like RMS.
Slashdot Mirror
> The X-Windowing-System has come with xmag virtually for ever.
xmag is also quite inferior to the magnifier thingy that comes with
Windows 98.
However, X also has the ability to set a desktop any size you like
and then zoom in to view part of it at 640x480 with Ctrl-Alt-+,
which with a large monitor should provide quite good magnification.
This is much closer to the functionality that the Windows magnifier
provides, because you can easily pan around. If it were possible to
zoom in to even lower resolutions (320x200 comes to mind) it would
be better. (Maybe it _is_ possible to do that, by writing a custom
modeline for those resolutions, but the GUI config tools for X that
ship with Mandrake don't make it easy to set that up.)
Really, xmag ought to be improved (or replaced) to be resizeable
and support a follow-the-mouse-cursor mode.
Unfortunately, it doesn't seem to work with the regular cursor
movement keys, only with the keypad, but I find it necessary to
remap the keypad to emulate the regular cursor keys, on the grounds
that otherwise most X apps don't support the keypad properly for
cursor movement (which is *way* more important to me than mouse
movement). (Just using the non-pad cursor keys is unacceptable
because they're arranged bady.)
I suppose I could map things so that the regular arrows emulate
the pad arrows... I never had a reason before (having no use
for the non-cursor functionality of the pad), but this might be
a good enough reason.
Do the pad home/end/pgup/pgdn keys do anything meaningful with
the shift+numlock keyboard-mouse mode turned on?
> there -will- be a future for anyone who fell in love with BeOS
I believe the most meaningful future contribution of BeOS is the
set of ideas it has made available to the developer community. The
BeOS was *innovative*, innovative in ways no current major OS is
innovative.
We're already seeing ideas from the BeOS incorporated into other
systems. Filesystem journaling is a good example. The new (yet
to be released) MS filesystem is another, perhaps better, example.
Have you noticed people in the OSS community talking about threading
lately, and about making things thread-safe? Heck, have you seen
processors for sale with hyperthreading technology? Thank Be for
demonstrating the performance advantages of a heavily threaded
system. These are not pie in the sky but real, concrete benefits
that we have because of the BeOS.
There's another thing current OSes need to learn from BeOS; Linux
has started, but there's a long way to go: better on-the-fly
hardware recognition. It's nothing to take a hard drive with the
BeOS on it out of one computer and stick it in another computer,
turn it on, and just start using it. No fooling with driver CDs,
no reconfiguring X, no "New Hardware Found" dialog boxes, just
turn it on and use it. (This is also part of what makes BeOS so
easy to install.) Kudzu and HardDrake are starting to be pretty
impressive, but the BeOS still has them beat on this point, so
there's room to grow.
Okay, so it's missing some things like a working security model,
the ability to set global color preferences, and (since it's not
1998 anymore) support for modern hardware. But we can learn from
the things it's *not* missing.
I still want a versioned filesystem. ITS had it in the days of
yore; VMS has had it for a good while too. C'mon, somebody get
on the stick and put it in a modern OS.
> Do not meddle in the affairs of LeoDV, for he is subtle and quick
> to anger.
Never leave a live dragon or an angry ent out of your calculations.
> the lawyers are more like the Ringwraiths
I think Microsoft has more than nine lawyers. If they're more
formidable than ordinary orcs, perhaps they're Uruk Hai or trolls
or something. The Nazgul metaphor must be reserved for something
there are only a handful of, like totally satisfied Microsoft
customers or rock-solid secure Windows servers. Say, where's Eowyn?
> but we should really be debating how we get this right on an
> OSS platform
For starters we get ourselves a VHLL or three and stop writing
everything in %$@! C and C++. There will still be bugs, of course.
But I'm getting *really* tired of hearing about newly discovered
buffer overruns; that Can't Happen to an application written in any
decently modern language.
> Dude I hope this is a typo...
No, just standard understatement for effect.
> But that is precisely why extensions exist. So that you don't
> have to have all of those features installed.
The problem is not with the fact that the extensions are optional.
The problem is that the install process for them is far too
cumbersome for people who want most of them. I had to spend
over half an hour installing them at home (dialup), and half
that time at work. That's time I had to be actively involved
with the install process, because there were regular dialogs to
answer new install links to click. It ought to be possible to
check some checkboxes for which ones to install, click a couple
of buttons and then *do something else* while it happens.
> Personally, I think that's how it should stay. Keep the browser
> lean and let people modify it to install whichever extensions
> they want.
Fine, but I should be able to install a bunch of them at one go.
This nonsense about spending fourty-five minutes installing thirty
individual extensions one-by-one, answering 2-3 dialog boxes for
*each* of them (and needing to read these boxes, since the question
about which directory to install into has the yes/no backwards for
about half of them) is for the birds (harhar).
I want one button I can click to install them all, and one dialog
to answer for *all* of them whether to install in my profile or
globally. Then if I want I can disable two or three of the
misguided ones (like the one that turns the alt attribute for
img elements into an abbrev tag like in Netscape 4).
People who only want one or two extensions should still be able
to install them individually, of course.
> Perl 4 had Unicode support the day it was ported to plan9
Who on earth needed Unicode support in the days of Perl4? Unicode
didn't gain buzzword status until a couple of years ago, right about
the same time as XML. I'm still not sure *why* we need it, other
than that of course everything has to support it these days to be
considered modern and hip. Bah. XML at least is occasionally useful.
Regarding screen sizes: the same people who complain about small type will insist they need a small screen, that 18" is too large. They cannot coherently explain why smaller things are easier to see, but a large percentage of them insist on it, including almost everyone who wears bifocals. Many of them use 14" viewables and don't totally maximize the browser window.
In order to compensate for the small size of their screens, they will cut the resolution to 640x480, and I'm convinced they'd set it smaller than that if Windows would let them. Then on top of that they will only effectively use about half of the screen, saying that they can't see the top part, the bottom part, or whatever. The implication is that if this is your target market, your site needs to be usable at about 600x200. Whee. If you plan it right from the start, it is entirely possible to design a site that looks decent at that size and still scales and looks okay at much higher resolutions, but with bitmap-type graphics there are limits; it's going to look quite stretched at 1600x1200.
There are two ways around this: one is to make your graphics scale, and the other is to build your site mostly out of text, maybe using graphics for borders and backgrounds and stuff.
Making graphics scale *properly* means vector graphics. When you run across a vector graphics format with wide browser support please let me know, as I'd be very interested. As a kludge, you can use bitmapped graphics (PNG or whatever) and assign relative widths. I have done this once or twice (width="100%" in my case) in a pinch, but if you try it you will immediately see the problem. Depending on the graphic it might be okay for some things, but it's definitely not a general solution to the whole issue.
At this time, my recommendation for sites that need to scale well to different resolutions is to make heavy use of text and style sheets and use a few strategically-placed graphics to spruce things up without interfering too much with scalability. For example, a background graphic that can be tiled will accomodate different resolutions fairly well. Narrow borders that scale or repeat in one direction are another fine example. A medium-sized logo that can be centered at the top, above the rest of the content, may be designed such that it looks fine surrounded by varying amounts of whitespace. And so on.
As far as text, use the relative size attributes to make some text larger than the rest as necessary, but don't fix hard sizes, as some legacy browsers[1] then won't let the user scale the text; with relative sizes the browser will pick up and use the user's base font size.
Some of your layout problems can be lessened by use of alpha-channel transparency. This doesn't work with all browsers, though; it works with all browsers based on Mozilla.org code, recent versions of Opera, and possibly certain others, but not for example with old versions of Netscape. There is a kludge to make it work with some versions of MSIE, but this fails sometimes depending on the user's settings and in any case will not work with old versions of MSIE. If you are interested in pursuing alpha channel stuff, I have some examples up here: http://cgi.galion.lib.oh.us/test/ Especially see GPL-plus.html (for an example how how it helps layouts) and png-alpha/png-alpha-demo-hacked.html (for a demo of the transparency channel itself). There is also a really cool demo here, but that one has not been hacked to work with MSIE.
One final piece of advice: test your site at at least three resolutions, including 640x480 and 1280x1024.
[1] Notably Netscape 4 and all versions of MSIE.
> All I know is I had a giant plug with four prongs on it.
Oh, those. Incidentally, you can make a standard phone line work
with only two of those four wires. (This is still true with an
RJ12 connector; you only need two of the wires, for a voice line.)
If you think it's bad trying to keep RJ12 and RJ45 straight, you
ought to have to deal with the *other* kinds of modular connectors.
RJ12 has four wires and RJ45 has eight, but did you know, there are
two different kinds with six wires, differing only by the placement
of the little clip thingy that holds them in the socket? The one
with the centered clip is RJsomething (I forget the number, but it's
between 12 and 45); the off-center one is called MMJ or DEC423. I
have a crosspinned inline coupler for this type... and a real,
non-historical use for it.
> any sentence challenging English usage or pronunciation that ends
> in a preposition needs revisiting
"at" in that sentence is not functioning as a preposition; it is
functioning as the complementary part of the verb. Besides, the
rule "never end a sentence with a preposition" is significantly
oversimplistic; the correct rule is that the words in a prepositional
phrase must be kept together, in this order: the preposition first,
followed by any standard attributive adjectives modifying the object,
followed by the object itself, followed by any additional modifiers
(such as modifying phrases or clauses). The occurrance of other
words, not part of any prepositional phrase, that in other
circumstances might be used as prepositions, is irrelevant.
> now beat it, kid, before **I** hit you over the head with my PK-88!
Hush, or I'll hunt you down like a wumpus and make you program
a Quake workalike in CoBOL.
> You seem to be forgetting that a critical part of the
> laminating process is extreme heat.
The big, fast, convenient laminators work that way. The cheapo
ones just use two rolls of clear contact paper (one top and one
bottom), rollers, and a hand crank. Should be fine.
* Think before you act. If it would annoy *you*, chances are
it may annoy the guy nextdoor, also.
* Keep the noise down. If you must have sound from your PC or
stereo, get headphones. If you need to carry on a conversation,
go to the person, rather than yelling across the room.
* Don't do anything you see done in a Dilbert cartoon.
* Shower or bathe at least once a week whether you need it or not.
There may be a handful of other things peculiar to the environment,
but I'm certain that you can get 95% of the way there with basic
everyday common sense.
This is an excellent opportunity for web-standards zealots to roll
out the FUD. "If your website uses plugins, you could be sued for
patent infringement, forced to pay back royalties, and put out of
business! Quickly, convert all your websites to XHTML/CSS and be
safe from the patent lawyers!"
> actually IE loads faster, Firebird browses faster
IE loads significant parts of itself at system start time, so it
has less to load when you launch it. Mozilla (SeaMonkey) used to
have an equivalent feature, but circa 1.0 it was tworked (to improve
stability; apparently there were a number of memory-handling bugs
that could be worked around by forcing a reload of everything) and
hasn't really worked properly since. Firebird has AFAIK never had
this feature; I don't know whether there are any plans to add it.
> I have not tried Mozilla on a sub Ghz machine, nor do I intend to.
> Some of us like to keep current with the times... my THREE previous
> machines were faster than 1ghz.
I recently (this past spring) upgraded. Previously I was using a
PentiumII/233, and now it's a Pentium4/2.somethingGHz with the 800
MHz FSB. I've noticed very significant performance improvements
for some applications -- ifile, for example -- and for games.
Mozilla, however, performs pretty much exactly the same. This is
probably because I had my PII/233 maxed out on 512MB of RAM, which
was enough to make Mozilla happy. Browsing the web uses a lot more
RAM than CPU. Also, my internet connection hasn't changed, nor
has my hard drive changed. My RAM is now DDR, supposedly faster
than the SDRAM I had before, but that doesn't seem to make as much
difference.
Now, a while back when I added RAM to my PII/233 to bring it from
192 to 512MB, I did notice major performance improvements for
Mozilla at that time. I conclude that RAM is a major factor for
Mozilla performance, and CPU isn't. There are other major factors
besides RAM; your internet connection obviously would be one;
CPU speed doesn't seem to be, though.
> Agreed. I couldn't tell any difference either.
The difference is, FireBird has a smaller memory footprint than
SeaMonkey. As such, if you have a marginal amount of RAM, FireBird
may fit without swapping where SeaMonkey doesn't. This will
result in a significant perceived performance improvement. If you
have plenty of RAM, however, SeaMonkey will use more of it (and maybe
take a bit longer to start the application initially), but apart from
that the rendering speed will be the same -- they are both using
Gecko, after all.
> Also, since I keep mozilla open all the time it isn't too
> important how long it takes to load.
Indeed. I haven't restarted my browser since the big power outage,
and that's pretty typical for me. Often I'll leave it running from
the day I install it until the day I install the next version. It's
too much of a pain to simultaneously bring everything I was doing to
a close all at once. The ability to bookmark tabsets helps somewhat
with that, but you still lose your place, any sessions, forms you
were filling out, and so on. I'm tempted to skip 1.5 and wait for
the next version, just because I don't feel like restarting my
browser again yet...
> The sad, sad news is that Firebird and Thunderbird will not made :-(
> it into 1.5
If you've been testing Firebird and Thunderbird this is good news.
They're not ready. Firebird is getting there, and hopefully will
be ready to replace Navigator by 1.6 time, but SeaMonkey really
can't be put out to pasture if only Navigator has been adequately
replaced. Thunderbird... well, it still needs a lot of work.
Also, Sunbird needs to be working before SeaMonkey can be dropped.
Actually, Firebird has most of the features Navigator has, *if* you
install a metric tonne of Extensions. (This is a major issue,
however; it takes considerably longer and *many* times more
clicking to download and install all those extensions as compared
to just downloading and installing the entire SeaMonkey suite. A
solution needs to be worked out wherein many extensions can be
downloaded and installed in one go.) Even with all of the
extensions, though, FB is still missing a couple of very major
features, like the DOM inspector (which is dogfood, or should
be -- it's painful to do any work on themes without it; it's quite
handy for web development also).
> Once Gimp is updated to use the newer version of GTK, I'll :o)
> be very happy though
The new version of Gimp (1.3) uses the new GTK. You will, however,
want to keep your existing Gimp 1.2 install around also. The new
Gimp has some cool features, but at this point it's still also
missing some things that are present in 1.2.
> One MORE reason why HP-UX is the most GODAWFUL WORST *NIX
Are you certain you don't have it confused with XENIX?
> Remember "social engineering" only works on people with social
> skills! We read BOFH articles in the same way as "HOW-TO" documents!
User: I'm having a little trouble starting up Notepad...
BOFH: That's because we're standardising everyone on two text
editors, to maintain consistency across the network. We
upgraded the Windows systems from Notepad to EDLIN last
night during overnight processing.
User: But I don't know how to use EDLIN!
BOFH: Whose fault is that?
User: You said two text editors. What's the other one?
BOFH: The Unix systems all have sed. You wouldn't believe
the whining we got from the vim nerds and Emacs geeks,
but they'll get over it.
> at this point i would venture to say that most average people
> would be lost without OE and O when it comes to email.
Bunk. I install Pegasus Mail for end users -- users who don't know
how to copy and paste and are afraid to learn, people who don't
understand multitasking, don't know what the taskbar is for,
believe they need to "x out" of one app to get to something else --
and they don't have nearly as much trouble with Pegasus mail as
the people have who are using Outlook -- and I'm not talking
about trouble with viruses; I'm talking about people being able
to figure out how to read their mail. Pegasus has been that easy
to use since before there was Outlook. Yes, it has advanced
features, but you have to go digging through the menus and the
preferences dialogs to find them.
Now, some really powerful mailreaders aren't appropriate for end
users. The one I use (Gnus) is certainly not right for them. But
mailreaders exist that are easier to use than Outlook and better.
And free, unless you think like RMS.