I can get away with telling this, because I majored in math.
A doctor, a lawyer, and a mathematician were discussing the relative merits of having a wife versus a mistress. The doctor was talking about stress levels and extolling the physiological benefits of emotional stability; for health reasons, he said, it's better to have a wife. The lawyer was talking about legal issues and liabilities and settlements, and he claimed that it was preferable to have a mistress.
The mathematician said it was better to have both, because that way, when the wife thinks you're with the mistress, and the mistress thinks you're with the wife, you can get some time to yourself and do mathematics.
> But it's not completely unreasonable. The prototype just has to show proof > of concept. The real implementation doesn't have to directly use anything > from the prototype.
How on earth do you propose prototyping a proof-of-concept for running undetectibly on Unix *or* for altering votes on a Unix-based system *or* for any of the other things this app supposedly did... using VB?
Sure, you could use VB to slap together a movie-style dialog box that says "How many votes do you want to flip", accepts the third password you type, then when you hit the go button shows an animation of a skull and crossbones, makes odd sound effects, and doesn't actually *do* anything... but any programmer who would call that a prototype or a proof-of-concept wouldn't have the foggiest notion how to actually write the software in question.
Re:While this is great for open source advocates..
on
TheOpenCD 2.0 Released
·
· Score: 1
> The problem is that 99.9% don't care about source code.
*I* don't want the source code if it's written in a language I don't grok, such as C or C++, which covers over 60% of the software I use. (If the source code is written in lisp or Perl, that's another matter; in that case I do want it, certainly.)
So don't explain it in terms of source code, if the person you're explaining it to won't care about that. You explain it that way to programmers, not to end users.
Explain it in terms of its other advantages, such as being more configurable or whatever, or explain it in terms of the development model ("Open-Source software is software that many programmers all around the world work on improving, instead of just one company") or something. It's worthless to tell an end user "YOU can get the source code." RMS only advocates telling them that because he's so deep into the IT world that the "end users" he works with are at least power users if not programmers who happen to not actively develop the particular software in question. If he had to deal on a regular basis with real technophobes, the kind of people who want to write down a list of steps on paper when you show them how to copy and paste, which they promptly forget, he would know that it's meaningless to talk about source code with those people. You put that in the fine print of the license they won't read and instead talk about the software on its merits.
> Can be very close to keeping one's head buried in the sand.
If you go RTFA, you won't think so anymore. That article is one of the most humorously pathetically bad pieces I've read in quite a significant while. It tries to pass itself off as investigative journalism, but the style is all totally wrong for that. (The word "alleged" doesn't occur once in the whole thing, for example, a dead giveaway that it's not the mainstream press article it wants to be.) The most hilarious thing, though, the thing that had me rolling on the floor, was when the article stated flatly that VB5 was used to prototype a program that would run undetectably on unix-based systems.
If you don't have silica gel, rice is also a pretty descent dessicant. (This is why people put rice in salt shakers -- it keeps the salt from caking in humid weather.) The problem with dessicants is, they only work until they're saturated -- then you have to somehow demoisturize them, which can be tricky because they really cling to the moisture -- that's why they're useful as dessicants in the first place.
> I can't think a fifty degree garage would be problematic
Fifty would be fine -- that's room temperature, or vanishingly close to it.
But he's talking about leaving this in an unattached (to the house), unheated garage, year-round. In the wintertime, it could get down to twenty below. That's *not* room temperature. In fact, it's coat-wearing weather. Also garages tend to heat up quite a lot in the summer (probably because there's nothing between the roof, which the sun heats directly, and the main interior -- houses are protected by their attics) -- if it's 100 outside, it could get to 130 inside the garage easily, and that also is definitely not room temperature (and, with the CPU producing heat, the inside of the PC case would be warmer yet).
> He used to use DEC VT-100s. Those things lasted 10+ years easily (except the > keyboards), but in a PC, he needs new fans every six months or so and a > new hard drive every year or so.
The VT100s don't have fans or hard drives in them, which explains why they're okay. They're really little more than a monitor and keyboard, with a little interface circuitry. (We use VT510s at work.) But VTs by themselves won't do much -- they've got to be hooked up to a computer (usually either a Vax or an Alpha). I suppose the computer was kept indoors with his old setup, with CAT4 or somesuch running out to the VTs. If he wants the same sort of setup with PCs, you could run a long KVM extension cable from the garage in to an indoor PC. (This won't help for the OP's backup solution though.)
> I'm in the process of building a new home recording studio. When I originally > moved into my new (very old) house, I decided that in the interests of > conserving energy, I would replace most of the incandescent lighting fixtures > or lamps in my home with fluorescent fixtures or compact fluorescent > replacement lamps in those fixtures which could not easily be replaced.
There are two places where fluorescent lighting really doesn't belong: a recording studio is one of them. (The other is a room with a lot of CRTs, because the flicker can reach epilepsy-inducing levels.)
> Unfortunately, these fixtures are creating a massive amount of radio > frequency interference in my home. The worst culprits seem to be the > dimmable fluorescent fixtures in my living room.
Yeah, they do that.
> Barring replacing all my fixtures and lamps with conventional incandescents,
Okay, don't replace all of them then. How about just the ones in the most strategic places, such as near the recording equipment? Oh, and the ones you don't replace, make sure they're on a separate electrical circuit from the electronic equipment. (Lighting should generally be on a separate circuit from outlets anyway... but fluoresent lighting should definitely be on its own circuit.)
> can anyone point me in the direction of alternatives? Is it possible that > the decreasing quality of most home goods has led to a decreasing quality > in fluorescent ballast systems that are much more noisy from an RFI > standpoint? Some of these fluo's are so noisy, they even emit audible sound!
Fluorescent lights have always been that way. I've never seen a fluorescent light that didn't emit audible sound. Different ones have different pitches, and some are more ignorable than others, but they're all quite audible.
> It's gotten so bad that I can't even play an electric guitar without > turning off all the non-incandescent lighting in my house, which pretty > much limits me to playing and recording during daylight hours (when I'm > supposed to be out making money)."
It is conceivable that not all of the interference is coming through the air (and therefore largely unblockable without blocking the light) -- some of it may be coming to the sound equipment via the power cables. It might be worth a try to put the sound equipment on a power-conditioning UPS, the kind with sine-wave output. This is expensive, significantly more expensive than incandescent lamps (though, depending on how integrated the fixtures are into the building, probably still cheaper than replacing all your fixtures), but it *might* work, or at least help. Still, I'd get at least one incandescent lamp (a desk lamp will do in a pinch) for right next to the sound equipment, and switch off any fluorescent light that's right there.
> December 1st they are talking about DateTime which is an extremely useful
DateTime is more than just useful -- it's *essential*. I have no idea why it isn't in the core distribution, but it's on my list of things that have to be installed on any new computer before it's ready for use. Seriously, when was the last time you wrote a program of any significant size that *didn't* have to deal with dates and times? I use DateTime in *most* of my programs. (No, don't talk to me about localtime and gmtime; those are okay if all you want to know is what time it is right now (err, when the script was started, which is usually right now), but if you have to actually *do* anything with dates or times, you need something better.)
The really great thing about DateTime is the large collection of related modules -- DateTime::Format::MySQL, DateTime::Format::Mail, and so on and so forth, one for just about every date/time format known to man. This makes it really easy to link various things together, converting dates from whatever format into DateTime objects for internal use (and for working with durations and stuff) and then formatting it in whatever format you need as necessary.
DateTime and the modules associated with it have saved me hours of fiddling around with dates, time and again, in so many different projects I've lost count many times over, adding up to hundreds of hours I'm sure -- and I'm not even really a developer per se. (I do some development, but I also do support and training and network stuff and you name it; development is just a small part of my job really.)
> Both arguments however are flawed when it comes to experienced users.
Experienced users know how to change their launcher/link/shortcut/whatever to add the command-line argument that suppresses the splash screen. The splash screen exists, however, to notify *inexperienced* users that the app has, in fact, launched. If there were no splash screen, many users would continue clicking and clicking and end up with eight copies of the same application running. (I have seen this happen MANY times.)
The problem is not the existence of the splash screen; the problem is that it is not subject to good user-interface guidelines, such as: let the user resize things; let the user choose which window is in front; let the user minimize things that get in the way -- and so on and so forth. If the splash screen behaved itself like a normal window, you wouldn't dislike it so much.
> Yes there is, at least for non-resizeable dialogs. How do you know which > items to stretch by what amount? Scaling up everything proportionally will > look very bad most of the time.
Preferably things should be spaced out, but at the very least, the user ought to be allowed to resize it larger, leaving extra blank space at the right and bottom. This is important because the designer never has exactly the same setup as all the users, and so on some systems the dialog box content will be clipped -- i.e., it won't fit. On others there will be wasted space. The user should *ALWAYS* be able to choose to resize it. The minute you step away from the default settings, you find out which cretinous application designers assumed nobody ever changes any settings. You can see this with badly-written apps on any platform: Windows, Mac, Gnome, KDE, wherever. I've even seen XUL apps with this problem, and I have *no* idea why Gecko, which definitely ought to know better, allows XUL windows to be unresizeable.
Non-resizeable dialogs are in the same category with non-resizeable and non-minimizeable and non-movable always-on-top splashscreens: there's absolutely no good reason why the window manager can't allow the user to override these things. Sure, give the apps a way to tell the wm that they want to be always-on-top and free of the usual controls, but let the user override it. We've got several mostly-useless keys on the keyboard that could be co-opted for this: let the user hold the window key and press the pause key, or whatever, to restore the missing controls (border for resizing, buttons for minimizing, and so forth). ratpoison goes to the extreme and just makes all windows maximized all the time, but it's possible to give some control back to the user without going quite that far.
> 2 minutes worth of power to cleanly shutdown. UPS is ok to weather the power > shortage, a battery inside the power supply would allow for clean shutdown.
It shouldn't even need to be enough to shutdown -- all it needs is to dump the RAM and processor state (register contents and such) to a designated area on the hard drive (or flash RAM dedicated to this purpose, or whatever) from which the BIOS firmware can restore everything when power comes back. The OS would not even need to know the power was ever out, except to fix the system time.
> What's with all the talk about OOo being so damned slow? I find it > incredibly responsive on my machines.
A couple of points. First, it's usually fairly responsive once it's started (unless you exceed RAM and the OS has to swap it in and out, which can easily happen on older systems that haven't got as much RAM as could be desired), but these people are talking about startup time. Second, startup time for some reason is in my experience better on Linux than on Windows. I don't know why, and maybe it's my imagination, or maybe it's a result of the hardware I've seen it running on in each respective OS. There is the QuickStarter thing for Windows, but that is (or was, as of a version or two ago, at any rate) sufficiently buggy that depending on usage patterns it may be more trouble than it's worth -- and it doesn't help at all on systems with inadequate RAM.
Speaking of RAM: the amount of RAM is *way* more important for good OOo performance than the CPU speed. The system requirements for 1.1.3 say 64MB of RAM, but it will run much more smoothly with twice that.
No, the way splash screens are usually implemented is evil. (Granted, the OOo splashscreen falls into that category.)
> They pop up in front of you,
This part is okay, since you just clicked on the launcher/icon/whatever to start the app, you expect it to pop up. Even apps without splash screens do that. However...
> disabling you from doing what you were doing,
Yes, that's evil. There is a combination of factors responsible for this... * Splash screens usually set themselves to always-on-top, which they
shouldn't do, because it's needless and annoying. (However, the window
manager is also at fault for *allowing* an app to make itself always on
top; that should be the user's decision.) This is evil. * Splash screens usually don't have window decorations, so they're hard to
minimize or move or whatever. (Again, the window manager is also partly
at fault for allowing windows to be created without the proper controls.)
This too is evil.
> and don't allow you to start doing what you started the app for.
Done right, the splashscreen should go away immediately once the app is ready to go. Granted, reducing app startup time is also a good thing. But splash screens are not inherently evil per se; it's the way they're usually implemented that's evil.
> Does this mean that conservatives are opposed to Jeb Bush's systematic > disenfranchising of minority voters in Florida?
We would be, very much so, if it had happened, or if there was any reason to believe it had happened. However, it's been well-documented (and demonstrated clearly in the last gubernatorial election) that Florida in general and minority voters in Florida in particular have, since the 2000 election, leaned to the right. The minority voters turned out for Bush in Florida. (This is not quite as odd as if the minority voters in, say, Washington DC had turned Republican. Most of the minority voters in Florida are hispanic, and a lot of them are Cuban, and as a demographic they have never been very solidly Democrat. Think of them as swing voters.)
There are a collection of theories as to why: discontent at the way the Democratic party conducted themselves after the 2000 election, approval of the way the whole hurricane mess was handled, views on foreign policy, a combination of factors,... we don't know which of these theories are or are not correct, but it's clear that in 2002 and 2004 the minority voters in Florida turned out for the Republicans. This was evident in 2002, and it was evident in 2004 before the election, from the poll data. It surprised no one who was paying any attention. Kerry was betting on picking up other states that Gore had lost to make up for it.
The only *actual* evidence I have seen that could even be construed as an indication of anything improper here is the fact that in certain counties, people didn't vote the way they were registered. News flash: swing voters *frequently* don't vote the way they're registered, and if there are major current events influencing their vote they often swing like that in rather significant numbers -- and this swing to the right in Florida is absolutely no surprise; Florida has been swinging right for about four years now and was already significantly to the right of party registration numbers in 2002. The only way you can see anything improper there is if you believe that it's improper or unlikely for a voter to change his mind and vote for the other party, but in fact that happens all the time and has never been regarded as improper by the law -- and if a law was passed that *made* that improper, that should scare you to death.
> And to the "challengers" the GOP paid to prevent Ohio residents from > voting if they looked like they were likely to vote Democrat?
Actually, I was opposed to that, but not to the same degree. Several points are worth making here:
* At least on paper, the goal was to prevent voting fraud (i.e., people
voting more than once, voting on behalf of dead people, and so on and
so forth). Yeah, I know, whether that was the whole motivation is
suspect.
* Nobody was prevented from voting. At worst, voters were issued
provisional ballots that, if the election was close enough that they
could have an impact on the outcome, would be counted if it was
determined that everything was above-board and non-fraudulent. The
treatment of every case was observed by representatives of both parties,
as are *all* election matters in Ohio, and it is worth noting that the
people doing the complaining are mostly not the representatives of the
Democratic party who observed these matters on behalf of the boards of
elections. They are other people, poking into what they didn't observe.
Conspiracy theorists mostly.
> And to Kenneth Blackwell making sure that heavily Democratic areas of Ohio > don't have enough voting machines? And to the Nevada Republicans shredding > Democratic voter registration forms?
These allegations I had not even *heard* before, despite fairly regular reading of slashdot, which leans rather far to the left on average (check th
> You seriously think the republicans wouldn't be doing the exact same > thing if they were down by 40-odd votes?
If they ever do, they'll lose the respect of half their core demographic. Conservatives (well, many of us anyhow) fundamentally don't think that way. We think in terms of what's the right thing to do, *not* in terms of what thing can we do that will obtain the outcome we want. (Philosophers call these two ways of thinking about ethics "deontological" versus "teleological" theories of obligation. It's fascinating to read up on, because they're completely different paradigms -- and you really can't understand someone coming from the other perspective unless you're aware of this issue. I had absolutely no understanding of liberals until I understood how a teleological theory of obligation works.)
I'm sure there are people in the Republican party who *would* do such a thing, but I could not in good conscience vote for person whom I thought would behave that way. (Yeah, I mostly believe in voting for the lesser evil, but there is a point where the evil is too great to endorse, and an ethical system where the end can justify the means is over the line as far as I'm concerned.)
You have to understand how conservatives think on this issue: it's *wrong* for a candidate to deliberately undermine the election process just to get himself elected. It's not just a bad decision or a poor choice; it's fundamentally wicked. Liberals don't see things that way, which is why Democrats can get away with pulling such schenanighans without alienating their support base. If a Republican tried it, even if he *did* win that election, he'd never win the next one, because his reputation with the conservatives who voted for him the first time would be irreparable marred.
This is true for most things, but when it comes to the meanings of words, the whole point of words is to communicate with people -- so what people think a word means does indeed very much matter. Torvalds is not an executive in any normal sense of the word executive.
> Exactly what do you need even non-root shell access for?
The biggest thing a web developer needs privs for is installing modules off the CPAN. If you have to bug the admin every time, it wastes his time and yours -- especially if you need to keep certain frequently-updated modules up-to-date. (Even if not, an active site can need a module installed every couple of days.)
It's possible for the admin to set things up so that a non-root user can run CPAN.pm and install into a directory specific to that user, but this is not the case out of the box and in any case is not usually good enough for installing modules that CGI scripts need to use, since you usually can't install to a location the CGI scripts will have access to. Even if you can, you can't install XS modules this way -- you have to be root to do that.
> Now if I could just find a firefox extention to remove the subdomains > from slashdot.org automatically so I don't have to deal with any of > these terrible color schemes (it.slashdot.org for example).
Edit->Preferences Click the Fonts & Colors button, and make sure Always Use My Colors is checked. This works not just for subdomains at slashdot, but also for every other page on the web too. You'll never have to look at a garish colour scheme on the web again. Set your colours to whatever you like, and keep them that way. I haven't gone snowblind from looking web pages with Evil Blinding White Backgrounds in years, because I have my colors set to #FFE6BC on #294D4A, a much easier-on-the-eyes combination.
>...and before Tezuka, the story was called Hamlet.
I don't know the Tezuka thing, but the Lion King is nothing whatsoever like Hamlet. In TLK, the protagonist is a bumbling careless child who is forced to grow up and take responsibility at the end; in Hamlet the protagonist is a careful schemer from the beginning who feigns madness, uses psychology to assure himself of the villain's guilt, carefully ponders whether to do what he's about to do at every step, and survives a plot on his life by cunning, playing along with it knowingly and turning it against the perpetrators, in sharp contrast to the Lion King, who is manipulated by the plotters, flees, and survives only because of the assassins' laziness.
There are similar elements to the story, sure. For example, the villain is a relative who also killed the protagonist's father -- but that much of the story goes back *way* before Hamlet. (Numerous times in the history of the Roman Empire it actually happened, and it wasn't original then.)
> When will people learn that no story is 100% original?
This is true, but the Lion King is more like Aladin than it is like Hamlet.
Slashdot Mirror
It sounds like a Fremen trap to me.
I can get away with telling this, because I majored in math.
A doctor, a lawyer, and a mathematician were discussing the relative merits
of having a wife versus a mistress. The doctor was talking about stress
levels and extolling the physiological benefits of emotional stability; for
health reasons, he said, it's better to have a wife. The lawyer was talking
about legal issues and liabilities and settlements, and he claimed that it
was preferable to have a mistress.
The mathematician said it was better to have both, because that way, when the
wife thinks you're with the mistress, and the mistress thinks you're with the
wife, you can get some time to yourself and do mathematics.
> But it's not completely unreasonable. The prototype just has to show proof
> of concept. The real implementation doesn't have to directly use anything
> from the prototype.
How on earth do you propose prototyping a proof-of-concept for running
undetectibly on Unix *or* for altering votes on a Unix-based system *or*
for any of the other things this app supposedly did... using VB?
Sure, you could use VB to slap together a movie-style dialog box that says
"How many votes do you want to flip", accepts the third password you type,
then when you hit the go button shows an animation of a skull and crossbones,
makes odd sound effects, and doesn't actually *do* anything... but any
programmer who would call that a prototype or a proof-of-concept wouldn't
have the foggiest notion how to actually write the software in question.
> The problem is that 99.9% don't care about source code.
*I* don't want the source code if it's written in a language I don't grok,
such as C or C++, which covers over 60% of the software I use. (If the
source code is written in lisp or Perl, that's another matter; in that
case I do want it, certainly.)
So don't explain it in terms of source code, if the person you're explaining
it to won't care about that. You explain it that way to programmers, not
to end users.
Explain it in terms of its other advantages, such as being more configurable
or whatever, or explain it in terms of the development model ("Open-Source
software is software that many programmers all around the world work on
improving, instead of just one company") or something. It's worthless to
tell an end user "YOU can get the source code." RMS only advocates telling
them that because he's so deep into the IT world that the "end users" he
works with are at least power users if not programmers who happen to not
actively develop the particular software in question. If he had to deal
on a regular basis with real technophobes, the kind of people who want to
write down a list of steps on paper when you show them how to copy and paste,
which they promptly forget, he would know that it's meaningless to talk about
source code with those people. You put that in the fine print of the license
they won't read and instead talk about the software on its merits.
> Can be very close to keeping one's head buried in the sand.
If you go RTFA, you won't think so anymore. That article is one of the most
humorously pathetically bad pieces I've read in quite a significant while. It
tries to pass itself off as investigative journalism, but the style is all
totally wrong for that. (The word "alleged" doesn't occur once in the whole
thing, for example, a dead giveaway that it's not the mainstream press article
it wants to be.) The most hilarious thing, though, the thing that had me
rolling on the floor, was when the article stated flatly that VB5 was used
to prototype a program that would run undetectably on unix-based systems.
> But you'd prefer to allow government officials to keep their positions
> even if they actually cheated in the elections?
That's not what the OP was saying. He was saying he desparately hopes that
said alleged cheating didn't actually happen. I tend to agree.
If you don't have silica gel, rice is also a pretty descent dessicant. (This
is why people put rice in salt shakers -- it keeps the salt from caking in
humid weather.) The problem with dessicants is, they only work until they're
saturated -- then you have to somehow demoisturize them, which can be tricky
because they really cling to the moisture -- that's why they're useful as
dessicants in the first place.
> I can't think a fifty degree garage would be problematic
Fifty would be fine -- that's room temperature, or vanishingly close to it.
But he's talking about leaving this in an unattached (to the house),
unheated garage, year-round. In the wintertime, it could get down to twenty
below. That's *not* room temperature. In fact, it's coat-wearing weather.
Also garages tend to heat up quite a lot in the summer (probably because
there's nothing between the roof, which the sun heats directly, and the
main interior -- houses are protected by their attics) -- if it's 100
outside, it could get to 130 inside the garage easily, and that also is
definitely not room temperature (and, with the CPU producing heat, the
inside of the PC case would be warmer yet).
> He used to use DEC VT-100s. Those things lasted 10+ years easily (except the
> keyboards), but in a PC, he needs new fans every six months or so and a
> new hard drive every year or so.
The VT100s don't have fans or hard drives in them, which explains why they're
okay. They're really little more than a monitor and keyboard, with a little
interface circuitry. (We use VT510s at work.) But VTs by themselves won't
do much -- they've got to be hooked up to a computer (usually either a Vax
or an Alpha). I suppose the computer was kept indoors with his old setup,
with CAT4 or somesuch running out to the VTs. If he wants the same sort
of setup with PCs, you could run a long KVM extension cable from the garage
in to an indoor PC. (This won't help for the OP's backup solution though.)
> I'm in the process of building a new home recording studio. When I originally
> moved into my new (very old) house, I decided that in the interests of
> conserving energy, I would replace most of the incandescent lighting fixtures
> or lamps in my home with fluorescent fixtures or compact fluorescent
> replacement lamps in those fixtures which could not easily be replaced.
There are two places where fluorescent lighting really doesn't belong: a
recording studio is one of them. (The other is a room with a lot of CRTs,
because the flicker can reach epilepsy-inducing levels.)
> Unfortunately, these fixtures are creating a massive amount of radio
> frequency interference in my home. The worst culprits seem to be the
> dimmable fluorescent fixtures in my living room.
Yeah, they do that.
> Barring replacing all my fixtures and lamps with conventional incandescents,
Okay, don't replace all of them then. How about just the ones in the most
strategic places, such as near the recording equipment? Oh, and the ones
you don't replace, make sure they're on a separate electrical circuit from
the electronic equipment. (Lighting should generally be on a separate
circuit from outlets anyway... but fluoresent lighting should definitely
be on its own circuit.)
> can anyone point me in the direction of alternatives? Is it possible that
> the decreasing quality of most home goods has led to a decreasing quality
> in fluorescent ballast systems that are much more noisy from an RFI
> standpoint? Some of these fluo's are so noisy, they even emit audible sound!
Fluorescent lights have always been that way. I've never seen a fluorescent
light that didn't emit audible sound. Different ones have different pitches,
and some are more ignorable than others, but they're all quite audible.
> It's gotten so bad that I can't even play an electric guitar without
> turning off all the non-incandescent lighting in my house, which pretty
> much limits me to playing and recording during daylight hours (when I'm
> supposed to be out making money)."
It is conceivable that not all of the interference is coming through the
air (and therefore largely unblockable without blocking the light) -- some
of it may be coming to the sound equipment via the power cables. It might
be worth a try to put the sound equipment on a power-conditioning UPS, the
kind with sine-wave output. This is expensive, significantly more expensive
than incandescent lamps (though, depending on how integrated the fixtures
are into the building, probably still cheaper than replacing all your
fixtures), but it *might* work, or at least help. Still, I'd get at least
one incandescent lamp (a desk lamp will do in a pinch) for right next to
the sound equipment, and switch off any fluorescent light that's right there.
> December 1st they are talking about DateTime which is an extremely useful
DateTime is more than just useful -- it's *essential*. I have no idea why it
isn't in the core distribution, but it's on my list of things that have to be
installed on any new computer before it's ready for use. Seriously, when was
the last time you wrote a program of any significant size that *didn't* have
to deal with dates and times? I use DateTime in *most* of my programs. (No,
don't talk to me about localtime and gmtime; those are okay if all you want
to know is what time it is right now (err, when the script was started, which
is usually right now), but if you have to actually *do* anything with dates
or times, you need something better.)
The really great thing about DateTime is the large collection of related
modules -- DateTime::Format::MySQL, DateTime::Format::Mail, and so on and so
forth, one for just about every date/time format known to man. This makes it
really easy to link various things together, converting dates from whatever
format into DateTime objects for internal use (and for working with durations
and stuff) and then formatting it in whatever format you need as necessary.
DateTime and the modules associated with it have saved me hours of fiddling
around with dates, time and again, in so many different projects I've lost
count many times over, adding up to hundreds of hours I'm sure -- and I'm
not even really a developer per se. (I do some development, but I also do
support and training and network stuff and you name it; development is just
a small part of my job really.)
> Both arguments however are flawed when it comes to experienced users.
Experienced users know how to change their launcher/link/shortcut/whatever
to add the command-line argument that suppresses the splash screen. The
splash screen exists, however, to notify *inexperienced* users that the
app has, in fact, launched. If there were no splash screen, many users
would continue clicking and clicking and end up with eight copies of the
same application running. (I have seen this happen MANY times.)
The problem is not the existence of the splash screen; the problem is that
it is not subject to good user-interface guidelines, such as: let the user
resize things; let the user choose which window is in front; let the user
minimize things that get in the way -- and so on and so forth. If the splash
screen behaved itself like a normal window, you wouldn't dislike it so much.
> Yes there is, at least for non-resizeable dialogs. How do you know which
> items to stretch by what amount? Scaling up everything proportionally will
> look very bad most of the time.
Preferably things should be spaced out, but at the very least, the user ought
to be allowed to resize it larger, leaving extra blank space at the right and
bottom. This is important because the designer never has exactly the same
setup as all the users, and so on some systems the dialog box content will
be clipped -- i.e., it won't fit. On others there will be wasted space.
The user should *ALWAYS* be able to choose to resize it. The minute you
step away from the default settings, you find out which cretinous application
designers assumed nobody ever changes any settings. You can see this with
badly-written apps on any platform: Windows, Mac, Gnome, KDE, wherever.
I've even seen XUL apps with this problem, and I have *no* idea why Gecko,
which definitely ought to know better, allows XUL windows to be unresizeable.
> Or how about non-resizable dialogs
Non-resizeable dialogs are in the same category with non-resizeable and
non-minimizeable and non-movable always-on-top splashscreens: there's
absolutely no good reason why the window manager can't allow the user to
override these things. Sure, give the apps a way to tell the wm that they
want to be always-on-top and free of the usual controls, but let the user
override it. We've got several mostly-useless keys on the keyboard that
could be co-opted for this: let the user hold the window key and press the
pause key, or whatever, to restore the missing controls (border for resizing,
buttons for minimizing, and so forth). ratpoison goes to the extreme and
just makes all windows maximized all the time, but it's possible to give some
control back to the user without going quite that far.
> 2 minutes worth of power to cleanly shutdown. UPS is ok to weather the power
> shortage, a battery inside the power supply would allow for clean shutdown.
It shouldn't even need to be enough to shutdown -- all it needs is to dump the
RAM and processor state (register contents and such) to a designated area on
the hard drive (or flash RAM dedicated to this purpose, or whatever) from which
the BIOS firmware can restore everything when power comes back. The OS would
not even need to know the power was ever out, except to fix the system time.
> What's with all the talk about OOo being so damned slow? I find it
> incredibly responsive on my machines.
A couple of points. First, it's usually fairly responsive once it's started
(unless you exceed RAM and the OS has to swap it in and out, which can easily
happen on older systems that haven't got as much RAM as could be desired),
but these people are talking about startup time. Second, startup time for
some reason is in my experience better on Linux than on Windows. I don't
know why, and maybe it's my imagination, or maybe it's a result of the
hardware I've seen it running on in each respective OS. There is the
QuickStarter thing for Windows, but that is (or was, as of a version or two
ago, at any rate) sufficiently buggy that depending on usage patterns it may
be more trouble than it's worth -- and it doesn't help at all on systems with
inadequate RAM.
Speaking of RAM: the amount of RAM is *way* more important for good OOo
performance than the CPU speed. The system requirements for 1.1.3 say 64MB
of RAM, but it will run much more smoothly with twice that.
> Splash screens are evil!
No, the way splash screens are usually implemented is evil. (Granted, the
OOo splashscreen falls into that category.)
> They pop up in front of you,
This part is okay, since you just clicked on the launcher/icon/whatever to
start the app, you expect it to pop up. Even apps without splash screens
do that. However...
> disabling you from doing what you were doing,
Yes, that's evil. There is a combination of factors responsible for this...
* Splash screens usually set themselves to always-on-top, which they
shouldn't do, because it's needless and annoying. (However, the window
manager is also at fault for *allowing* an app to make itself always on
top; that should be the user's decision.) This is evil.
* Splash screens usually don't have window decorations, so they're hard to
minimize or move or whatever. (Again, the window manager is also partly
at fault for allowing windows to be created without the proper controls.)
This too is evil.
> and don't allow you to start doing what you started the app for.
Done right, the splashscreen should go away immediately once the app is
ready to go. Granted, reducing app startup time is also a good thing.
But splash screens are not inherently evil per se; it's the way they're
usually implemented that's evil.
> Does this mean that conservatives are opposed to Jeb Bush's systematic
... we don't know which of these theories are or
> disenfranchising of minority voters in Florida?
We would be, very much so, if it had happened, or if there was any reason to
believe it had happened. However, it's been well-documented (and demonstrated
clearly in the last gubernatorial election) that Florida in general and
minority voters in Florida in particular have, since the 2000 election, leaned
to the right. The minority voters turned out for Bush in Florida. (This is
not quite as odd as if the minority voters in, say, Washington DC had turned
Republican. Most of the minority voters in Florida are hispanic, and a lot
of them are Cuban, and as a demographic they have never been very solidly
Democrat. Think of them as swing voters.)
There are a collection of theories as to why: discontent at the way the
Democratic party conducted themselves after the 2000 election, approval of
the way the whole hurricane mess was handled, views on foreign policy, a
combination of factors,
are not correct, but it's clear that in 2002 and 2004 the minority voters
in Florida turned out for the Republicans. This was evident in 2002, and it
was evident in 2004 before the election, from the poll data. It surprised
no one who was paying any attention. Kerry was betting on picking up other
states that Gore had lost to make up for it.
The only *actual* evidence I have seen that could even be construed as an
indication of anything improper here is the fact that in certain counties,
people didn't vote the way they were registered. News flash: swing voters
*frequently* don't vote the way they're registered, and if there are major
current events influencing their vote they often swing like that in rather
significant numbers -- and this swing to the right in Florida is absolutely
no surprise; Florida has been swinging right for about four years now and
was already significantly to the right of party registration numbers in 2002.
The only way you can see anything improper there is if you believe that
it's improper or unlikely for a voter to change his mind and vote for
the other party, but in fact that happens all the time and has never been
regarded as improper by the law -- and if a law was passed that *made* that
improper, that should scare you to death.
> And to the "challengers" the GOP paid to prevent Ohio residents from
> voting if they looked like they were likely to vote Democrat?
Actually, I was opposed to that, but not to the same degree. Several
points are worth making here:
* At least on paper, the goal was to prevent voting fraud (i.e., people
voting more than once, voting on behalf of dead people, and so on and
so forth). Yeah, I know, whether that was the whole motivation is
suspect.
* Nobody was prevented from voting. At worst, voters were issued
provisional ballots that, if the election was close enough that they
could have an impact on the outcome, would be counted if it was
determined that everything was above-board and non-fraudulent. The
treatment of every case was observed by representatives of both parties,
as are *all* election matters in Ohio, and it is worth noting that the
people doing the complaining are mostly not the representatives of the
Democratic party who observed these matters on behalf of the boards of
elections. They are other people, poking into what they didn't observe.
Conspiracy theorists mostly.
> And to Kenneth Blackwell making sure that heavily Democratic areas of Ohio
> don't have enough voting machines? And to the Nevada Republicans shredding
> Democratic voter registration forms?
These allegations I had not even *heard* before, despite fairly regular
reading of slashdot, which leans rather far to the left on average (check th
> You seriously think the republicans wouldn't be doing the exact same
> thing if they were down by 40-odd votes?
If they ever do, they'll lose the respect of half their core demographic.
Conservatives (well, many of us anyhow) fundamentally don't think that way.
We think in terms of what's the right thing to do, *not* in terms of what
thing can we do that will obtain the outcome we want. (Philosophers call
these two ways of thinking about ethics "deontological" versus "teleological"
theories of obligation. It's fascinating to read up on, because they're
completely different paradigms -- and you really can't understand someone
coming from the other perspective unless you're aware of this issue. I had
absolutely no understanding of liberals until I understood how a teleological
theory of obligation works.)
I'm sure there are people in the Republican party who *would* do such a thing,
but I could not in good conscience vote for person whom I thought would behave
that way. (Yeah, I mostly believe in voting for the lesser evil, but there is
a point where the evil is too great to endorse, and an ethical system where
the end can justify the means is over the line as far as I'm concerned.)
You have to understand how conservatives think on this issue: it's *wrong*
for a candidate to deliberately undermine the election process just to get
himself elected. It's not just a bad decision or a poor choice; it's
fundamentally wicked. Liberals don't see things that way, which is why
Democrats can get away with pulling such schenanighans without alienating
their support base. If a Republican tried it, even if he *did* win that
election, he'd never win the next one, because his reputation with the
conservatives who voted for him the first time would be irreparable marred.
> What people think is very rarely the truth.
This is true for most things, but when it comes to the meanings of words, the
whole point of words is to communicate with people -- so what people think
a word means does indeed very much matter. Torvalds is not an executive in
any normal sense of the word executive.
> Yes, but IMs by their very nature tend to be a whole lot personal than emails.
Wow. You apparently live in a completely different universe than I do.
> "What about software upgrades? New Perl modules?" --Sorry, bub, installing
> and upgrading software is exactly what the sysadmin is there for.
You want the webmaster to bug the sysadmin every time a module needs installed
or upgraded? That can be several times a day!
> Exactly what do you need even non-root shell access for?
The biggest thing a web developer needs privs for is installing modules off the CPAN. If you have to bug the admin every time, it wastes his time and yours --
especially if you need to keep certain frequently-updated modules up-to-date.
(Even if not, an active site can need a module installed every couple of days.)
It's possible for the admin to set things up so that a non-root user can run
CPAN.pm and install into a directory specific to that user, but this is not
the case out of the box and in any case is not usually good enough for
installing modules that CGI scripts need to use, since you usually can't
install to a location the CGI scripts will have access to. Even if you can,
you can't install XS modules this way -- you have to be root to do that.
Thanks for the bookmarklet; that works great...
> Now if I could just find a firefox extention to remove the subdomains
> from slashdot.org automatically so I don't have to deal with any of
> these terrible color schemes (it.slashdot.org for example).
Edit->Preferences
Click the Fonts & Colors button, and make sure Always Use My Colors is checked.
This works not just for subdomains at slashdot, but also for every other page
on the web too. You'll never have to look at a garish colour scheme on the
web again. Set your colours to whatever you like, and keep them that way.
I haven't gone snowblind from looking web pages with Evil Blinding White
Backgrounds in years, because I have my colors set to #FFE6BC on #294D4A, a
much easier-on-the-eyes combination.
> ...and before Tezuka, the story was called Hamlet.
I don't know the Tezuka thing, but the Lion King is nothing whatsoever like
Hamlet. In TLK, the protagonist is a bumbling careless child who is forced
to grow up and take responsibility at the end; in Hamlet the protagonist is
a careful schemer from the beginning who feigns madness, uses psychology to
assure himself of the villain's guilt, carefully ponders whether to do what
he's about to do at every step, and survives a plot on his life by cunning,
playing along with it knowingly and turning it against the perpetrators, in
sharp contrast to the Lion King, who is manipulated by the plotters, flees,
and survives only because of the assassins' laziness.
There are similar elements to the story, sure. For example, the villain is
a relative who also killed the protagonist's father -- but that much of the
story goes back *way* before Hamlet. (Numerous times in the history of the
Roman Empire it actually happened, and it wasn't original then.)
> When will people learn that no story is 100% original?
This is true, but the Lion King is more like Aladin than it is like Hamlet.