> What was the last time anyone used the function keys that are on the numpad?
Just now. The problem with the new-style cursor keys added in the 101-key layout is that you can't reach them all at once. That makes them substantially less usable than the cursor-control keypad.
However, the last time I used the cursor keypad for entering numbers was going on a decade ago, before I learned to actually touchtype. If I want numbers, I'll type numbers. The keypad is for cursor control. I'm actually using a custom keyboard layout that prevents number lock from inadvertently getting turned on.
> you can't count on having a syntax highlighted editor all the time
Dude, if I've got a system that's such a fresh new install it hasn't even got a working Emacs yet, there's no WAY I'm going to be messing with SQL on it! That's like worrying about what CD is in the CD-ROM drive when you haven't even bolted down the motherboard to the tray yet.
With that said, I don't see the need for syntax highlighting *or* capitals for SQL. Maybe the SQL I work with just isn't as complicated as you guys, but I tend to just, you know, type it. Or else I use something like Class::DBI.
> SELECT * FROM tblWhatever a INNER JOIN tblYaddaYadda b ON a.ID = b.ID WHERE...
Umm, that's what the SHIFT key is for. If your SHIFT KEY is in a SENSIBLE location you can TYPE IN ALL CAPS as desired without even having a CAPSLOCK key on your keyboard. (My layout, for example, has no CAPSLOCK.)
Oh, and dude, lose the ntnHungarian. One-character sigils are one thing, but three-letter two-syllable prefixes like tbl are *entirely* too much.
> If you're an Emacs user, having the capslock key mapped to control is the > ONLY way to fly.
No, there is another. I have ctrl under my right pinky (home position). BTW, shift is under my left pinky (home position). I have an Avant keyboard, so my layout is as custom as I want it to be. It's *mostly* QWERTY, but I've made some very key[1] changes. Central to these changes is removing the need to hyperextend my pinkies on a constant basis.
Incidentally, there is no CAPSLOCK in my layout; if I want capital letters, I'll type capital letters using shift; that's part of what it's there for.
I do have a ctrl key in the bottom left, which I can use for Ctrl-C and Ctrl-X and Ctrl-V when my right hand is off the keyboard (e.g., on the mouse), but for most ctrl combinations I use the right ctrl, which as noted is a home position.
[1] I didn't even *notice* that pun until I hit Preview. Honest.
> But 20 is enough to start getting a better idea of whether or not something > works consistantly. > > "it worked on mine machine" is meaningless in the windows world.
Agreed, though I think you can start getting *some* idea at less than 20. (Not really at 1 or 2, though.) Administering six PCs is enough to let you weed out the stuff that works on about three quarters of all Windows systems and fails the other quarter of the time. I administer about a dozen Windows desktops, and I can tell you that PS/2, parallel, and serial ports may not be zooming fast, but they are absolutely the way to go for stuff that doesn't really need the speed, e.g. mice, keyboards, and most desktop printers. A dozen Windows systems is quite enough that I've seen USB devices fail to work entirely too often. My attitude has developed into, "USB: Just Say No".
> I wonder why the idea never caught on to have a standard, hardware interface > designed for home soldering enthusiasts (the port was designed to be physically > large enough to manipulate without special equipment).
We already have a standard for that. It's called RS232. Been around forever.
Yeah, but it's a software firewall, which is a really long way from ideal. Really you should put Windows systems behind an external firewall (that is, a firewall that doesn't run on Windows), such as a hardware firewall or a *nix box of some kind.
And that still won't protect you if you use Outlook, though as far as I am aware there hasn't been a new Outlook virus yet this month, so things could be looking up on that front.
> What we really need to do is figure out how to disable wireless phones in an > area about the size of a movie theatre or concert hall. Perhaps something > slightly less lethal than a shotgun.
The lethality of a shotgun depends on what kind of ammo you use. For example, if you load your shotgun with rock salt, it's a lot less lethal than if you use buck shot.
> I'm not going to claim the syntax looks good at first glance
The major objection to sendmail isn't syntax; it's security. sendmail is on the very short list of programs I disallow on my network for security reasons. Its security track record is every bit as bad as IIS, and the problem is a core problem with the philosophy of the developers: they patch specific vulnerabilities, but they don't have any interest in fixing the core design that _leads_ to all those vulnerabilities.
Fundamentally, sendmail runs as root while processing untrusted data arriving from the internet. That's a major fundamental security no-no. You just don't *do* that. Apache doesn't do that. proftpd doesn't do that. There's no *need* to do that, but sendmail does it anyway for arcane historical reasons.
> but then most perl programs look like line noise too
Now *you're* trolling. The only Perl programs that look like line noise are the ones that are deliberately obfuscated, like my signature.
I've known that hotmail occasionally loses mail since... well, since before they switched the hotmail servers over to NT, anyway. I wouldn't have expected it to change since then, particularly since it's a free service. In other news, Yahoo! mail occasionally has quite significant delays (several hours or more) when sending or receiving, and some messages can get delayed a lot more than others so that mail arrives out-of-order (which can be really weird if you're on a mailing list).
> All that said, the idea of having to use both keyboard and mouse for such > a fundamental operation is just so horrifyingly backwards and wrong
That's exactly why Ctrl-C and Ctrl-V is better: no need to take my hands off the keyboard. (No, I don't usually use the mouse for selection; the keyboard is much faster, once you learn to use the cursor keypad, which I did years ago, back in the days of DOS 3. I do have a mouse, so I can use Gimp:-)
> Now, I'll grant you that the current situation with regards to cutting and > pasting in X is badly designed. But I don't think copying windows is the way > to fix it. As I posted elsewhere, I think multiple buffers a la vi are the > answer.
Why not a kill ring, a la Emacs?
Yes, that's tongue-in-cheek; as an avid Emacs user[1], I've yet to discover a use for the kill ring beyond the top entry -- and as an avid X11 user, I've yet to discover a use for having more than one clipboard, unless you count confusing people as a "use". There are quite a lot of things I dislike about Windows, but the cliboard is *NOT* one of them; it works, it works right, it works in virtually all applications, and it works that way every single time. And you don't lose the clipboard contents when you close an app, either. If other aspects of Windows were as solid as the clipboard, I'd probably still be using it.
I'm not generally a big fan of copying Windows, but for this we should make an exception (or, think of it as copying Mac if you prefer, or the BeOS).
[1] In the "know and use Emacs lisp regularly" category. I have a rather
sizeable collection of custom lisp that automates a lot of stuff for
me, and hardly any of my keystrokes still have their default bindings.
I use Gnus for email and eshell for commandline stuff. If there were
a real use for the kill ring, one of the people on gnu.emacs.help would
have pointed it out to me. Kai probably would have posted a useful tip
about it, or something. Anyway, I'd know.
> Currently this is not the case. There are for example significant issues > with moving stuff back and forth between Mozilla and Emacs, under X11.
For the record, this thread motivated me to figure out the problem here. Apparently x-select-enable-clipboard has a default value of nil (hysterical raisins, no doubt), but to get the correct behavior it must be set to t.
Well, that solves my problem for Emacs, but Emacs wasn't the only app I ever had this problem with under X11, only the most important one. And it irks me that the default is the clearly wrong behavior.
> IIRC, what should happen is the primary selection (ctrl-c/ctrl-v) should > be seperate from secondary selection (select text, then middle click)
This creates serious problems when going from one application to another. IMO, there should only be *one* clipboard, and *all* applications and toolkits should support it. It is not necessary for the same keystroke (or mouse) combination to be used in all applications, but *whatever* keystrokes (or mousestrokes) are used for copy and paste in any given application should interoperate with the *same* clipboard used by all other applications.
Currently this is not the case. There are for example significant issues with moving stuff back and forth between Mozilla and Emacs, under X11. I don't have this problem with the same applications on Windows. (I have other problems under Windows, but they're not relevant to this thread.)
I don't know (or care) whether the problem is with X11's having multiple clipboards or whether the problem is with applications supporting them improperly; what I do know (and care about) is that it's not obvious how to copy and paste information between applications that use different toolkits, and that's bad.
The other annoying thing about highlight/middle-click copying and pasting is that it requires me to put my hand on the mouse. I should *NOT* have to put my hand on the mouse to copy and paste when I'm typing. All applications and toolkits SHOULD support a form of copying and pasting that can be done with the keyboard -- and it should use the same one clipboard as all other applications.
Sure, restoring a disk image occasionally might work okay for a single PC in a rental situation, but it would be a maintenance nightmare in a lab. With multiple PCs in the room and multiple people using each one per day (and on occasion multiple people per hour), you could end up hiring two or three full-time restore-monkeys just to reghost the suckers.
No, for a lab situation you want diskless thin clients that boot off the network thin client server, which is physically locked up and configured so that the clients don't have write access to the boot image at the host operating system level. (If you want them to be able to save settings and stuff, you give them home or user directories on a separate network file server and set up the client OS to mount those. And you set up quotas.)
One of the things that keeps me from being able to use sodipodi for more than a very short amount of time at once is that it violates a fundamental rule of accessibility and forces a certain background color on you. (The Linux version of OO.o used to do this (though the Windows version never did), but it's been fixed now.) Alas, Inkscape seems to have this same problem, forcing the entire image area to be blinding white even if it's transparent as far as SVG is concerned. This makes it basically unusable for me, as my eyes are far too sensitive to light to have a blinding white background on a large area of the screen extended periods. I go snowblind.
Can someone tell me where there's a setting in sodipodi or Inkscape (or any other freely available vector graphics editor) to get it to honor my system background color setting (any of them -- Qt, either version of GTK, or even Win32, as I could do vector graphics on the WinMe box in a pinch if it would solve this issue for me)? I'd like to do more with vector graphics, but as it stands I'm going to have to mostly stick with doing overly-large bitmapped graphics in Gimp and scaling them down to whatever size I need.
If she shows any interest at all in text-based games, introduce her to Inform. About a third of the people who learn Inform do so saying, "I'm not really a programmer, I just want to create this game I have in my head..." Some of them move on to other languages later.
Inform is really good for teaching object-oriented design, too. It was my third language, after GW-BASIC and Pascal, and it revolutionized the way I thought about certain kinds of programming problems.
Also, the Designer's Manual is really top-notch, especially the section on the object-oriented world model with the Ruins example game. It's a joy to read. Really excellent stuff. Recommended.
> There is nothing more nauseating than someone who quotes MP at length, > trying to be funny. It's basically the sure sign of someone who just > isn't funny at all. Al Gore probably does it at parties.
Quoting M. Python _at length_ isn't funny; it's stupid. Quoting a short but well-timed and well-chosen excerpt occasionally can, however, be quite funny.
> Speaking as a Brit, I found Chicken Run hugely disappointing.
I didn't like it either, but I'm not much of a movie buff. (I started several years ago compiling a list of the top-ten best overall movies of all time, and I still haven't managed to come up with ten movies worthy of being on the list.) Some people I know who are less picky liked it better.
> This whole Adams worshipping strikes the wrong note with me. I mean, the > guy was great, but like the rest of us, he had his occassional shit ideas. > I've read the early draft of the "Salmon of Doubt". He worked over and over > on scripts to bring them up to par.
Agreed, wholeheartedly. Good books come from good authors; great books come from decent authors with the tenacity and humility to rewrite and revise at length, repeatedly, ruthlessly scrapping whatever doesn't work and going back and redoing it again until they get it right. I know Tolkien is in the latter category; his books (well, ones anybody besides avid Tolkien fans have heard about) all went through lengthy and repeated revision processes. (The extreme example of this is the Silmarillion, but LOTR was revised over quite a few years before publication as well.) This is why his books are so incredibly great; if they weren't, he rewrote them until they were. I suspect Adams is in this category as well. TSOD is nowhere *near* the level of Adams' other work, and it's probably because he hadn't revised it enough times yet.
However, THHGTG had already been worked over repeatedly, first as a radio series and then as a book for the British market and yet again when it was revised for the American market. By this point, the materiel was in pretty good shape, overall. Less needed to be done, hopefully, than needed to be (and wasn't) for the Salmon of Doubt.
Whether the filmmakers ruin it, of course, is an open question that remains to be seen. They've managed to ruin quite a few films that *ought* to have been great, had great source material, and were just screwed up by inane decisions, bad directing, bad casting, and a lack of understanding of the author's vision. What they did to Great Expectations is simply inexcusable, for example. OTOH, they did right by Hamlet; casting Mel Gibson in that one was dead-on right, and they mostly resisted the urge to mess with the script, and it came out beautifully.
> After all, not even Peter Jackson, with his vision, scope, funding and love > of the books could silence the complaints following the rings trilogy.
Maybe that's because of the way he handled the materiel: "Hmmm... I'm not sure why these books are some of the most popular ever, the guy wasn't as great of a writer as I think I am. I think I'll throw out half the action and almost all of the dialog and rewrite it basically from scratch, using only certain parts I particularly liked. These characters, I don't like their attitudes and tone; I think I'll change them around a bit, turn these guys into comic relief, make this really noble guy an evil meanie, get rid of one of the most beloved characters altogether,... I wouldn't want for anyone to confuse my movies with the books, or think they actually follow the same story, or anything like that."
There is, however, an important difference between LOTR and HHGG. HHGG isn't trying to be serious. It's supercilious by design. It's internally consistent in a wide assortment of areas even _within_ the books, and that's fine, because of the type of work it is. It's not meant to be taken seriously. So making small changes, as long as they're in the same _spirit_ as the original, doesn't fundamentally ruin the story.
LOTR is different from that. It *is* meant to be taken seriously. There are light moments, yes, but the overwhelming tone of the books is *serious*, even solemn. The author went far out of his way to maintain consistency not only internally within LOTR but also with the companion works (especially There and Back Again and also the Silmarillion). For crying out loud, LOTR has a sizeable set of appendices that flesh out the backstory, so that the reader can gain a full understanding of the world in which the books are set. This is a whole different type of book from HHGG, and it ought to have been treated rather differently. Jackson made sweeping gratuitous changes, changes that served no purpose and significantly weakened the storyline. Perhaps his worst error was attempting to cram a series of six[1] rather lengthy and involved books into only three movies. Any *one* of the six books in LOTR contains more material than the entire five-book HHGG trilogy.
I don't mean in terms of the number of words or pages (though they are rather longer individually than the individual books of HHGG), but rather in terms of the amount of storyline and dialog. HHGG spends a lot of its words going on about things that wouldn't translate well into movie form, such as humorous pseudotechnical explanations of various pieces of Adams' special brand of physics. Tolkien doesn't explain his world model directly (except in the appendices); it comes out as backstory that you can piece together from bits and pieces of comments made by the characters and things that take place. Almost all of the text of LOTR is either stuff happening or the characters' talking. It would all translate fairly straightforwardly into movie form.
I could have forgiven Jackson for leaving out some of the lengthier passages of dialog (e.g., most of the council at Rivendell), but he left out entire major action scenes, including *the* most significant action scene from the first book. If HHGG were done the way Jackson did LOTR, the movie wouldn't feature Zaphod Beeblebrox, and instead of Arther and Ford being tossed out the airlock and rescued from space by the Heart of Gold they would get off the Vogon ship when it arrives at Magrathea. And yes, if he does that, Douglas Adams' fans will want to lynch the directors. HHGG can handle a bit of inconsistency due to its inherent silliness, but there have to be limits.
> For us ignorant Brits, wthat's that in Gas Mark?
Not sure about Gas Mark (Is that a guy named mark who eats too much cabbage with his beans?), but broiling is a type of cooking. It can be done in an oven (using a broiler pan), or over a fire (using a grill), but the key thing about broiling is that there are lots of open slits under the food so that the grease can drip down through, away from the food. You end up with food that's considerably less greasy. Ribs are often broiled.
"Broil" is not really a specific temperature pre se, but a lot of electric ovens do have a "Broil" setting, and it's usually just shy of "Clean". HTH.HAND.
Slashdot Mirror
> What was the last time anyone used the function keys that are on the numpad?
Just now. The problem with the new-style cursor keys added in the 101-key
layout is that you can't reach them all at once. That makes them substantially
less usable than the cursor-control keypad.
However, the last time I used the cursor keypad for entering numbers was going
on a decade ago, before I learned to actually touchtype. If I want numbers,
I'll type numbers. The keypad is for cursor control. I'm actually using a
custom keyboard layout that prevents number lock from inadvertently getting
turned on.
> you can't count on having a syntax highlighted editor all the time
Dude, if I've got a system that's such a fresh new install it hasn't even got
a working Emacs yet, there's no WAY I'm going to be messing with SQL on it!
That's like worrying about what CD is in the CD-ROM drive when you haven't
even bolted down the motherboard to the tray yet.
With that said, I don't see the need for syntax highlighting *or* capitals for
SQL. Maybe the SQL I work with just isn't as complicated as you guys, but I
tend to just, you know, type it. Or else I use something like Class::DBI.
> SELECT * FROM tblWhatever a INNER JOIN tblYaddaYadda b ON a.ID = b.ID WHERE ...
Umm, that's what the SHIFT key is for. If your SHIFT KEY is in a SENSIBLE
location you can TYPE IN ALL CAPS as desired without even having a CAPSLOCK
key on your keyboard. (My layout, for example, has no CAPSLOCK.)
Oh, and dude, lose the ntnHungarian. One-character sigils are one thing, but
three-letter two-syllable prefixes like tbl are *entirely* too much.
> If you're an Emacs user, having the capslock key mapped to control is the
> ONLY way to fly.
No, there is another. I have ctrl under my right pinky (home position). BTW,
shift is under my left pinky (home position). I have an Avant keyboard, so my
layout is as custom as I want it to be. It's *mostly* QWERTY, but I've made
some very key[1] changes. Central to these changes is removing the need to
hyperextend my pinkies on a constant basis.
Incidentally, there is no CAPSLOCK in my layout; if I want capital letters,
I'll type capital letters using shift; that's part of what it's there for.
I do have a ctrl key in the bottom left, which I can use for Ctrl-C and Ctrl-X
and Ctrl-V when my right hand is off the keyboard (e.g., on the mouse), but for
most ctrl combinations I use the right ctrl, which as noted is a home position.
[1] I didn't even *notice* that pun until I hit Preview. Honest.
Hie thee to xulplanet.com and install the Preferences Toolbar.
> But 20 is enough to start getting a better idea of whether or not something
> works consistantly.
>
> "it worked on mine machine" is meaningless in the windows world.
Agreed, though I think you can start getting *some* idea at less than 20. (Not
really at 1 or 2, though.) Administering six PCs is enough to let you weed out
the stuff that works on about three quarters of all Windows systems and fails
the other quarter of the time. I administer about a dozen Windows desktops, and
I can tell you that PS/2, parallel, and serial ports may not be zooming fast,
but they are absolutely the way to go for stuff that doesn't really need the
speed, e.g. mice, keyboards, and most desktop printers. A dozen Windows systems
is quite enough that I've seen USB devices fail to work entirely too often.
My attitude has developed into, "USB: Just Say No".
> comes from trying to support two architectures
Yeah, we know how much trouble that's been for BSD and Linux...
> I wonder why the idea never caught on to have a standard, hardware interface
> designed for home soldering enthusiasts (the port was designed to be physically
> large enough to manipulate without special equipment).
We already have a standard for that. It's called RS232. Been around forever.
> I thought XP comes with a build-in firewall.
Yeah, but it's a software firewall, which is a really long way from ideal.
Really you should put Windows systems behind an external firewall (that is,
a firewall that doesn't run on Windows), such as a hardware firewall or a
*nix box of some kind.
And that still won't protect you if you use Outlook, though as far as I am
aware there hasn't been a new Outlook virus yet this month, so things could
be looking up on that front.
> What we really need to do is figure out how to disable wireless phones in an
> area about the size of a movie theatre or concert hall. Perhaps something
> slightly less lethal than a shotgun.
The lethality of a shotgun depends on what kind of ammo you use. For example,
if you load your shotgun with rock salt, it's a lot less lethal than if you
use buck shot.
> It's the only MTA that has never had a security lapse.
The only *major* one perhaps...
> (actually, Courier might not have had one either, but who runs Courier?)
There are also other less-well-known options that don't have security problems.
Generally they also don't have tons of features, granted.
*Eventually* I want to write an entire mail-handling system in Perl. I've got
a working POP3 server. I want to do IMAP next...
> Perl programs ... are deliberately obfuscated, like my signature.
Uh, like my *former* signature, which I replaced with something else now.
> I'm not going to claim the syntax looks good at first glance
The major objection to sendmail isn't syntax; it's security. sendmail is on
the very short list of programs I disallow on my network for security reasons.
Its security track record is every bit as bad as IIS, and the problem is a
core problem with the philosophy of the developers: they patch specific
vulnerabilities, but they don't have any interest in fixing the core design
that _leads_ to all those vulnerabilities.
Fundamentally, sendmail runs as root while processing untrusted data arriving
from the internet. That's a major fundamental security no-no. You just don't
*do* that. Apache doesn't do that. proftpd doesn't do that. There's no
*need* to do that, but sendmail does it anyway for arcane historical reasons.
> but then most perl programs look like line noise too
Now *you're* trolling. The only Perl programs that look like line noise
are the ones that are deliberately obfuscated, like my signature.
I've known that hotmail occasionally loses mail since... well, since before
they switched the hotmail servers over to NT, anyway. I wouldn't have expected
it to change since then, particularly since it's a free service. In other news,
Yahoo! mail occasionally has quite significant delays (several hours or more)
when sending or receiving, and some messages can get delayed a lot more than
others so that mail arrives out-of-order (which can be really weird if you're
on a mailing list).
> All that said, the idea of having to use both keyboard and mouse for such
:-)
> a fundamental operation is just so horrifyingly backwards and wrong
That's exactly why Ctrl-C and Ctrl-V is better: no need to take my hands off
the keyboard. (No, I don't usually use the mouse for selection; the keyboard
is much faster, once you learn to use the cursor keypad, which I did years
ago, back in the days of DOS 3. I do have a mouse, so I can use Gimp
> Now, I'll grant you that the current situation with regards to cutting and
> pasting in X is badly designed. But I don't think copying windows is the way
> to fix it. As I posted elsewhere, I think multiple buffers a la vi are the
> answer.
Why not a kill ring, a la Emacs?
Yes, that's tongue-in-cheek; as an avid Emacs user[1], I've yet to discover a
use for the kill ring beyond the top entry -- and as an avid X11 user, I've
yet to discover a use for having more than one clipboard, unless you count
confusing people as a "use". There are quite a lot of things I dislike about
Windows, but the cliboard is *NOT* one of them; it works, it works right, it
works in virtually all applications, and it works that way every single time.
And you don't lose the clipboard contents when you close an app, either. If
other aspects of Windows were as solid as the clipboard, I'd probably still
be using it.
I'm not generally a big fan of copying Windows, but for this we should make
an exception (or, think of it as copying Mac if you prefer, or the BeOS).
[1] In the "know and use Emacs lisp regularly" category. I have a rather
sizeable collection of custom lisp that automates a lot of stuff for
me, and hardly any of my keystrokes still have their default bindings.
I use Gnus for email and eshell for commandline stuff. If there were
a real use for the kill ring, one of the people on gnu.emacs.help would
have pointed it out to me. Kai probably would have posted a useful tip
about it, or something. Anyway, I'd know.
> Currently this is not the case. There are for example significant issues
> with moving stuff back and forth between Mozilla and Emacs, under X11.
For the record, this thread motivated me to figure out the problem here.
Apparently x-select-enable-clipboard has a default value of nil (hysterical
raisins, no doubt), but to get the correct behavior it must be set to t.
Well, that solves my problem for Emacs, but Emacs wasn't the only app I ever
had this problem with under X11, only the most important one. And it irks me
that the default is the clearly wrong behavior.
> IIRC, what should happen is the primary selection (ctrl-c/ctrl-v) should
> be seperate from secondary selection (select text, then middle click)
This creates serious problems when going from one application to another.
IMO, there should only be *one* clipboard, and *all* applications and toolkits
should support it. It is not necessary for the same keystroke (or mouse)
combination to be used in all applications, but *whatever* keystrokes (or
mousestrokes) are used for copy and paste in any given application should
interoperate with the *same* clipboard used by all other applications.
Currently this is not the case. There are for example significant issues
with moving stuff back and forth between Mozilla and Emacs, under X11. I
don't have this problem with the same applications on Windows. (I have
other problems under Windows, but they're not relevant to this thread.)
I don't know (or care) whether the problem is with X11's having multiple
clipboards or whether the problem is with applications supporting them
improperly; what I do know (and care about) is that it's not obvious how
to copy and paste information between applications that use different
toolkits, and that's bad.
The other annoying thing about highlight/middle-click copying and pasting is
that it requires me to put my hand on the mouse. I should *NOT* have to put
my hand on the mouse to copy and paste when I'm typing. All applications
and toolkits SHOULD support a form of copying and pasting that can be done
with the keyboard -- and it should use the same one clipboard as all other
applications.
> Isn't that how you ran your labs?
Ack, I should certainly home not!
Sure, restoring a disk image occasionally might work okay for a single PC in
a rental situation, but it would be a maintenance nightmare in a lab. With
multiple PCs in the room and multiple people using each one per day (and on
occasion multiple people per hour), you could end up hiring two or three
full-time restore-monkeys just to reghost the suckers.
No, for a lab situation you want diskless thin clients that boot off the
network thin client server, which is physically locked up and configured so
that the clients don't have write access to the boot image at the host
operating system level. (If you want them to be able to save settings and
stuff, you give them home or user directories on a separate network file
server and set up the client OS to mount those. And you set up quotas.)
One of the things that keeps me from being able to use sodipodi for more than
a very short amount of time at once is that it violates a fundamental rule of
accessibility and forces a certain background color on you. (The Linux version
of OO.o used to do this (though the Windows version never did), but it's been
fixed now.) Alas, Inkscape seems to have this same problem, forcing the
entire image area to be blinding white even if it's transparent as far as
SVG is concerned. This makes it basically unusable for me, as my eyes are
far too sensitive to light to have a blinding white background on a large area
of the screen extended periods. I go snowblind.
Can someone tell me where there's a setting in sodipodi or Inkscape (or any
other freely available vector graphics editor) to get it to honor my system
background color setting (any of them -- Qt, either version of GTK, or even
Win32, as I could do vector graphics on the WinMe box in a pinch if it would
solve this issue for me)? I'd like to do more with vector graphics, but as
it stands I'm going to have to mostly stick with doing overly-large bitmapped
graphics in Gimp and scaling them down to whatever size I need.
If she shows any interest at all in text-based games, introduce her to Inform.
About a third of the people who learn Inform do so saying, "I'm not really a
programmer, I just want to create this game I have in my head..." Some of them
move on to other languages later.
Inform is really good for teaching object-oriented design, too. It was my
third language, after GW-BASIC and Pascal, and it revolutionized the way I
thought about certain kinds of programming problems.
Also, the Designer's Manual is really top-notch, especially the section on
the object-oriented world model with the Ruins example game. It's a joy to
read. Really excellent stuff. Recommended.
> There is nothing more nauseating than someone who quotes MP at length,
> trying to be funny. It's basically the sure sign of someone who just
> isn't funny at all. Al Gore probably does it at parties.
Quoting M. Python _at length_ isn't funny; it's stupid. Quoting a short but
well-timed and well-chosen excerpt occasionally can, however, be quite funny.
> Speaking as a Brit, I found Chicken Run hugely disappointing.
I didn't like it either, but I'm not much of a movie buff. (I started several
years ago compiling a list of the top-ten best overall movies of all time, and
I still haven't managed to come up with ten movies worthy of being on the list.)
Some people I know who are less picky liked it better.
> This whole Adams worshipping strikes the wrong note with me. I mean, the
> guy was great, but like the rest of us, he had his occassional shit ideas.
> I've read the early draft of the "Salmon of Doubt". He worked over and over
> on scripts to bring them up to par.
Agreed, wholeheartedly. Good books come from good authors; great books come
from decent authors with the tenacity and humility to rewrite and revise at
length, repeatedly, ruthlessly scrapping whatever doesn't work and going back
and redoing it again until they get it right. I know Tolkien is in the latter
category; his books (well, ones anybody besides avid Tolkien fans have heard
about) all went through lengthy and repeated revision processes. (The extreme
example of this is the Silmarillion, but LOTR was revised over quite a few
years before publication as well.) This is why his books are so incredibly
great; if they weren't, he rewrote them until they were. I suspect Adams is
in this category as well. TSOD is nowhere *near* the level of Adams' other
work, and it's probably because he hadn't revised it enough times yet.
However, THHGTG had already been worked over repeatedly, first as a radio
series and then as a book for the British market and yet again when it was
revised for the American market. By this point, the materiel was in pretty
good shape, overall. Less needed to be done, hopefully, than needed to be
(and wasn't) for the Salmon of Doubt.
Whether the filmmakers ruin it, of course, is an open question that remains
to be seen. They've managed to ruin quite a few films that *ought* to have
been great, had great source material, and were just screwed up by inane
decisions, bad directing, bad casting, and a lack of understanding of the
author's vision. What they did to Great Expectations is simply inexcusable,
for example. OTOH, they did right by Hamlet; casting Mel Gibson in that one
was dead-on right, and they mostly resisted the urge to mess with the script,
and it came out beautifully.
> After all, not even Peter Jackson, with his vision, scope, funding and love
... I wouldn't want for
> of the books could silence the complaints following the rings trilogy.
Maybe that's because of the way he handled the materiel: "Hmmm... I'm not
sure why these books are some of the most popular ever, the guy wasn't as
great of a writer as I think I am. I think I'll throw out half the action
and almost all of the dialog and rewrite it basically from scratch, using
only certain parts I particularly liked. These characters, I don't like
their attitudes and tone; I think I'll change them around a bit, turn these
guys into comic relief, make this really noble guy an evil meanie, get rid
of one of the most beloved characters altogether,
anyone to confuse my movies with the books, or think they actually follow
the same story, or anything like that."
There is, however, an important difference between LOTR and HHGG. HHGG isn't
trying to be serious. It's supercilious by design. It's internally consistent
in a wide assortment of areas even _within_ the books, and that's fine, because
of the type of work it is. It's not meant to be taken seriously. So making
small changes, as long as they're in the same _spirit_ as the original,
doesn't fundamentally ruin the story.
LOTR is different from that. It *is* meant to be taken seriously. There are
light moments, yes, but the overwhelming tone of the books is *serious*, even
solemn. The author went far out of his way to maintain consistency not only
internally within LOTR but also with the companion works (especially There
and Back Again and also the Silmarillion). For crying out loud, LOTR has a
sizeable set of appendices that flesh out the backstory, so that the reader
can gain a full understanding of the world in which the books are set. This
is a whole different type of book from HHGG, and it ought to have been
treated rather differently. Jackson made sweeping gratuitous changes,
changes that served no purpose and significantly weakened the storyline.
Perhaps his worst error was attempting to cram a series of six[1] rather
lengthy and involved books into only three movies. Any *one* of the six
books in LOTR contains more material than the entire five-book HHGG trilogy.
I don't mean in terms of the number of words or pages (though they are
rather longer individually than the individual books of HHGG), but rather
in terms of the amount of storyline and dialog. HHGG spends a lot of its
words going on about things that wouldn't translate well into movie form,
such as humorous pseudotechnical explanations of various pieces of Adams'
special brand of physics. Tolkien doesn't explain his world model directly
(except in the appendices); it comes out as backstory that you can piece
together from bits and pieces of comments made by the characters and things
that take place. Almost all of the text of LOTR is either stuff happening
or the characters' talking. It would all translate fairly straightforwardly
into movie form.
I could have forgiven Jackson for leaving out some of the lengthier passages
of dialog (e.g., most of the council at Rivendell), but he left out entire
major action scenes, including *the* most significant action scene from the
first book. If HHGG were done the way Jackson did LOTR, the movie wouldn't
feature Zaphod Beeblebrox, and instead of Arther and Ford being tossed out
the airlock and rescued from space by the Heart of Gold they would get off
the Vogon ship when it arrives at Magrathea. And yes, if he does that,
Douglas Adams' fans will want to lynch the directors. HHGG can handle a bit
of inconsistency due to its inherent silliness, but there have to be limits.
[1] Yes, six books. Published in three volumes.
> You come from the KiB camp, right?
Kids in Black?
> For us ignorant Brits, wthat's that in Gas Mark?
Not sure about Gas Mark (Is that a guy named mark who eats too much cabbage
with his beans?), but broiling is a type of cooking. It can be done in an
oven (using a broiler pan), or over a fire (using a grill), but the key thing
about broiling is that there are lots of open slits under the food so that
the grease can drip down through, away from the food. You end up with food
that's considerably less greasy. Ribs are often broiled.
"Broil" is not really a specific temperature pre se, but a lot of electric
ovens do have a "Broil" setting, and it's usually just shy of "Clean".
HTH.HAND.