Blocking outbound will still raise problems and confusion.
An example case: Getting my sister's stuff up and running at spymac.com. I had set up a spymac account for myself. Outgoing SMTP through them worked beautifully from Linux with GNUMail.app (as long as I specified my From address as @spymac.com - don't know the details, probably looks at the From and does some kind of pop-before-smtp authentication.)
Tried to get my sister online. Fetches mail from spymac just fine with POP3. Can't send a damn thing. We double-checked the settings. Triple-checked the settings.
Then, the truth dawned to me: My ISP (a local DSL provider) doesn't block outgoing SMTP. My sister's (free dial-up provider) does.
Normally, this wouldn't be a problem. All ISP's I've used here have pretty simple steps on how to get online. "Set your outgoing mail server to 'mail.whatever.fi'." No need to mess with the settings, just set that as the outgoing E-mail server in the mail proggy, or smarthost on a Linux MTA. The host would allow connections from ISP customers, refuse from the rest of the net, and trust whatever details have been put in the E-mail itself.
Not so with this ISP! Looking at their tech support, I should have needed the username at their own E-mail system ("Hey dad, do you remember your username at the ISP's E-mail system? Huh? You did know they give you an E-mail address, didn't you? Oh... never mind..."), which in turn would have given the address of the smtp server on the cluster. I didn't even try further. Let me guess: some form of SMTP authentication next, instead of trusting the IP address and the name on the envelope.
Let's see what kind of hideous bubblegum solution I come up with this weekend...
The point is, the more complicated you make sending the e-mail to whatever outgoing SMTP host you may need, the more complicated you make the life for some people who prefer not to use it. Of course, blocking it is still valid (I guess in this case - it was a free dialup ISP anyway, without such block spammers would use tons of throwaway accounts). I guess an acceptable solution won't be found until someone comes up with a set of better E-mail protocols that aren't spammer-friendly at all but that are flexible enough for normal users.
As others noted, this is definitely Perl HTML::Mason, which is one of the best web scripting environments I've ever worked in. An adequate comparision would be something like this: PHP's down-to-earth approach of mixing code and HTML, JSP's and Website Meta Language's ideas on how to separate them again if they need to be (code componentization and tag libraries), Perl as the scripting language, and Apache mod_perl to give it some speed (also works as CGI).
I'm just wishing to know how to turn the cool-looking error dumps off when they're viewed outside localhost =)
I'll turn 25 shortly. You know, it'd be extremely nice if within an year, I'd be officially too old to be a tech support person. Because every time I'm asked to do some technically challenging thing (like install something...) by a friend or relative, I get a serious headache.
"Could you get this thing working?" "Sorry dad, I'm too old to do tech support, I can only do it if the doctor says it's okay."
"Could you get this working?" "Why sure, little sister, if you let the me, the senile install guy, play a little bit of that OSX chess. You know, back when I was a kid, the Macintoshes all looked like bird nesting boxes, and these computer chess boards were really black-and-white (Blah blah blah...)"
well, they've picked the cross-platform compatibility as a goal right from the start - sticking to APIs that are not pain to port over.
Commercial game developers typically pick one or two platforms and then "optimize extensively" for those platforms. They want optimal, efficient and platform-aware programs, not academic-technical masterpieces. (Do a "grep -r goto *" in Quake source one day. Try to read the code without a noticeable eye bleeding. =)
Cube isn't developed with the goal of maximizing the profit from one or two platforms - it's made specifically with the other platforms and portability in mind. It apparently uses common cross-platform libraries like OpenGL and SDL, and many libraries that originate in *NIX but have been ported to other platforms (image libraries, Ogg Vorbis, etc). As an added bonus, being an open-source project, they can easily be open-source all the way and use whatever is already there.
Saying "What utter rubbish" is what causes the problem in first place. The definitions of kilo/mega/gigabytes are varying depending on who's speaking. And then there's the standard units, which are always the same. As long as people don't agree on something, and insist redefining the prefixes based on context, there will always be confusion.
I solved this problem this way: When I say "1 Gibibyte", it's 1073741824 bytes, and when I say "Gig", it's "About enough, but still too fucking much, to burn on a CD". =)
What a dork. Everyone knows that to kill the idle process you just run emacs for Windows.
Emacs? EMACS? You won't make a dent in any performance graph with that anymore, except maybe on startup (and every app needs resources at startup). 1990s called, they want their resource hog back!
Sure, on the 386 era Emacs was a resource eater, but these days, a proggy that eats like 15 megs of memory is nothing. And, hell, my window manager shows up in top(1) listing, but not XEmacs. I suppose emacsen aren't that hot anymore. For me, XEmacs starts up in seconds in both Linux and Windows.
I have to recommend OpenOffice.org as the new, improved cross-platform memory and proc eater. (Or maybe Mozilla with Java installed, though even that isn't much worse than Emacsen.)
But, really, who cares of the processor load/memory use unless it affects the performance?
Yeah, a Faraday cage. You know, this OptIn guy probably radiates spam waves so that without this kind of precautions all computers around him start spontaneously sending spam.
.org was supposed to designate a not-for-profit group.
Actually, strictly speaking,.org isn't "non-profit", it's "everyone else", as in "everyone who isn't a company, network operator, government site, military, or like." Which usually just happens to leave there non-profit organizations and private individuals.
Personally, I only give GLX's credit for SGI and the contributors. The only estabilished definition for GLX, of course, being the glue layer between OpenGL and X11, which is pretty widely used these days, coming right out of box with XFree86 and all.
Lesson number whatever: If you come up with a cool term, at least bother to google for it =)
For example, when Dennis Ritchie wrote the compiler for the BCPL language, he didn't actually use a computer. He scrawled the whole thing on a ream of paper, and had his secretary transcribe it.
Er, wasn't that rather the case with Niklaus Wirth and Pascal?
(The rumor has it that since initially there was no compiler for Pascal, Wirth wrote one in Pascal, then "interpreted" it in his head and "ran" it on paper. A secretary might indeed have been handy.)
MUDs, wherefrom the MMORPGs truly spring, have had addictive substances for quite a while now.
I think BatMUD has had tobacco addiction for as long as I can remember. Which would be something like since 1994-6 or something. Quite likely also earlier than that.
I don't know, but Wario Ware, Inc. covers pretty many genres and most of those microgames require pretty simple controls (some may need A+control, but a pretty large number is winnable with either A presses or just the controller).
And when you've got Wario Ware and Tetris, you've got everything! =)
BTW, speaking of content, remember when internet people were busy trying to generate it? You don't hear about that much anymore do you? AOL was the biggest culprit and, not surprisingly, the biggest loser as a result.
Well, what every expert said was "content matters, not the gimmics", the folks said "Yeah! Let's create some content!" with same results as you imagine.
Content still matters. However, it should indeed be remembered that "content" isn't created, it's grown. This was the fundamental error for some people.
Individual pieces of work aren't "content" as such - just like an almost completely empty house with a single chair in it isn't exactly well-decorated. Content is the whole thing.
In other words, you can't win if you think that your purpose is to throw some shit in and then go shouting "Yes, we have Content". Everyone should still be focusing individual bits and pieces rather than the whole picture.
Bah! You have no idea of true scambaiting art.This is art. A gripping tale of dark dark horror and greed. Guaranteed to spook you silly. Or not. =)
Seriously, anyone can get scammers to pose for silly photographs, but this tale was something completely extraordinary.
Okay, the 419eater scams are still funny enough, though... I particularly liked the one where the scammer tried to pass off photo from Vatican as a photo of their church =)
By the same token, a sculptor can take a three ring binder, some magic markers, and a broken keyboard and make a sculpture easily worth two thousand.
So this pppppowerbook is worth two thousand dollars? And cheap materials are worth tens of thousands?
Somehow, I'm imagining that the folks who came up with this way to trick scammers are laughing now, but wait 'til the scammer sells the pppppppowerbook to some modern art gallery or computer history museum. The scammers sure are smart! They can easily make zillions!
Completely untrue! Are you, perhaps, too too young to remember Taco Hell? =)
And ages ago, Slashdot was full of usual blog-type stuff like "Somerandomnickname writes, "Someone actually mentioned Linux somewhere". Wow." Or something along those lines. =)
Bloxs... Blosxom, despite its tricky name, can operate as a CGI script or it can generate static HTML and RSS files. A small and sweet app overall, nicely extendable too. All you need is Perl to run it locally.
Yup, Blosxom rules. I couldn't believe my eyes when I first saw it. Doesn't necessarily need a RDBMS (often these things only do MySQL! Horror!) to work, and can generate static pages, so it works on my web space provider (who also support CGI, but I preferred to do static pages).
Perhaps because the only good implementations out there right now are Vorbis reference implementations - especially on encoder side. It's interesting to see if anyone can come up with even better encoder. All that matters to me is that the files are compatible.
And more importantly, why didn't they take advantage of the chance to give it a better name than Vorbis? "aoTuV"? WTF?
Never question the Japanese, for they shall forevermore be shrouded in mystery and strangeness, and the other world simply cannot understand them at all times. =) Not that it's needed. They have strange naming ideas but good tech and software, and that's what matters.
Dunno about breaking the Message-ID part. I replaced the "groups" with "groups-beta" and it still found the message just as well - looks like the links won't get broken that easily when they roll this out. I do hope they make it still easy to find messages by Message-ID, though...
Suppose I had a Linux C++ application that I wanted to port to Windows and have a nice native GUI for it. Decided to use MinGW for this. There, I have all of the necessary headers and stuff for the low-level work. I can compile the base stuff easily on Windows, or even cross-compile them on Linux - basically most parts of the code that don't need UI or the obscure OS features.
Now, the UI work. I can get WTL from sourceforge, but how the heck do I get all of the headers and libraries it depends on (ATL and the others)? How difficult it is to obtain them? How do I get them to work nicely with g++? What kind of restrictions are there to using these libraries? Etc etc etc...
At this point, I'd probably just ditch it and go for wxwindows or something =)
Actually, I'd very much like to know this - I've heard some of the libraries are available through Microsoft with some restrictions. Anyone know better?
What happens when your portable CD player hits a bump every 0.75 seconds for minutes at a time? My portable CD player skips, even with a generous RAM buffer, but my GBA keeps on going, and that's why I carry a GBA and a flash cart loaded with 160 BPM eurobeat music when I jog.
Yeah, it skips sometimes; I just turn the buffering off! (Strange but true - when there's a lot of skipping, it seems to be *less* skippy with buffering on... Must be that the Area 51 technology from just a bit to the left from Rigel was cleverly disguised as cheap CD players and sent to the world with "Made in Taiwan" written on the back, or something. Nobody knew where to look from.)
As for jogging - music is overrated for jogging in my opinion. People are supposed to drop off all unnecessary junk and listen nothing but the sounds of their environment. The Nature. Or something like that. But yeah, if music is what you need for that, solid state memory rules. =)
God I guessed there would be a WarioWare title. What a better way to start the life of the "innovative two-display system" than to use it for something completely crazy? First, they laid down the ideas on how the dual-screen functionality could be used, and then, with WarioWare, they'll make a title that tosses the rules away for a while to do something unusual. Or at least that's what I'm definitely going to expect if it's going to be a genuine WarioWare title. I'm certainly hoping it will work that way =)
Slashdot Mirror
Blocking outbound will still raise problems and confusion.
An example case: Getting my sister's stuff up and running at spymac.com. I had set up a spymac account for myself. Outgoing SMTP through them worked beautifully from Linux with GNUMail.app (as long as I specified my From address as @spymac.com - don't know the details, probably looks at the From and does some kind of pop-before-smtp authentication.)
Tried to get my sister online. Fetches mail from spymac just fine with POP3. Can't send a damn thing. We double-checked the settings. Triple-checked the settings.
Then, the truth dawned to me: My ISP (a local DSL provider) doesn't block outgoing SMTP. My sister's (free dial-up provider) does.
Normally, this wouldn't be a problem. All ISP's I've used here have pretty simple steps on how to get online. "Set your outgoing mail server to 'mail.whatever.fi'." No need to mess with the settings, just set that as the outgoing E-mail server in the mail proggy, or smarthost on a Linux MTA. The host would allow connections from ISP customers, refuse from the rest of the net, and trust whatever details have been put in the E-mail itself.
Not so with this ISP! Looking at their tech support, I should have needed the username at their own E-mail system ("Hey dad, do you remember your username at the ISP's E-mail system? Huh? You did know they give you an E-mail address, didn't you? Oh... never mind..."), which in turn would have given the address of the smtp server on the cluster. I didn't even try further. Let me guess: some form of SMTP authentication next, instead of trusting the IP address and the name on the envelope.
Let's see what kind of hideous bubblegum solution I come up with this weekend...
The point is, the more complicated you make sending the e-mail to whatever outgoing SMTP host you may need, the more complicated you make the life for some people who prefer not to use it. Of course, blocking it is still valid (I guess in this case - it was a free dialup ISP anyway, without such block spammers would use tons of throwaway accounts). I guess an acceptable solution won't be found until someone comes up with a set of better E-mail protocols that aren't spammer-friendly at all but that are flexible enough for normal users.
As others noted, this is definitely Perl HTML::Mason, which is one of the best web scripting environments I've ever worked in. An adequate comparision would be something like this: PHP's down-to-earth approach of mixing code and HTML, JSP's and Website Meta Language's ideas on how to separate them again if they need to be (code componentization and tag libraries), Perl as the scripting language, and Apache mod_perl to give it some speed (also works as CGI).
I'm just wishing to know how to turn the cool-looking error dumps off when they're viewed outside localhost =)
I'll turn 25 shortly. You know, it'd be extremely nice if within an year, I'd be officially too old to be a tech support person. Because every time I'm asked to do some technically challenging thing (like install something...) by a friend or relative, I get a serious headache.
"Could you get this thing working?" "Sorry dad, I'm too old to do tech support, I can only do it if the doctor says it's okay."
"Could you get this working?" "Why sure, little sister, if you let the me, the senile install guy, play a little bit of that OSX chess. You know, back when I was a kid, the Macintoshes all looked like bird nesting boxes, and these computer chess boards were really black-and-white (Blah blah blah...)"
well, they've picked the cross-platform compatibility as a goal right from the start - sticking to APIs that are not pain to port over.
Commercial game developers typically pick one or two platforms and then "optimize extensively" for those platforms. They want optimal, efficient and platform-aware programs, not academic-technical masterpieces. (Do a "grep -r goto *" in Quake source one day. Try to read the code without a noticeable eye bleeding. =)
Cube isn't developed with the goal of maximizing the profit from one or two platforms - it's made specifically with the other platforms and portability in mind. It apparently uses common cross-platform libraries like OpenGL and SDL, and many libraries that originate in *NIX but have been ported to other platforms (image libraries, Ogg Vorbis, etc). As an added bonus, being an open-source project, they can easily be open-source all the way and use whatever is already there.
Saying "What utter rubbish" is what causes the problem in first place. The definitions of kilo/mega/gigabytes are varying depending on who's speaking. And then there's the standard units, which are always the same. As long as people don't agree on something, and insist redefining the prefixes based on context, there will always be confusion.
I solved this problem this way: When I say "1 Gibibyte", it's 1073741824 bytes, and when I say "Gig", it's "About enough, but still too fucking much, to burn on a CD". =)
Emacs? EMACS? You won't make a dent in any performance graph with that anymore, except maybe on startup (and every app needs resources at startup). 1990s called, they want their resource hog back!
Sure, on the 386 era Emacs was a resource eater, but these days, a proggy that eats like 15 megs of memory is nothing. And, hell, my window manager shows up in top(1) listing, but not XEmacs. I suppose emacsen aren't that hot anymore. For me, XEmacs starts up in seconds in both Linux and Windows.
I have to recommend OpenOffice.org as the new, improved cross-platform memory and proc eater. (Or maybe Mozilla with Java installed, though even that isn't much worse than Emacsen.)
But, really, who cares of the processor load/memory use unless it affects the performance?
Yeah, a Faraday cage. You know, this OptIn guy probably radiates spam waves so that without this kind of precautions all computers around him start spontaneously sending spam.
Actually, strictly speaking, .org isn't "non-profit", it's "everyone else", as in "everyone who isn't a company, network operator, government site, military, or like." Which usually just happens to leave there non-profit organizations and private individuals.
Personally, I only give GLX's credit for SGI and the contributors. The only estabilished definition for GLX, of course, being the glue layer between OpenGL and X11, which is pretty widely used these days, coming right out of box with XFree86 and all.
Lesson number whatever: If you come up with a cool term, at least bother to google for it =)
Er, wasn't that rather the case with Niklaus Wirth and Pascal?
(The rumor has it that since initially there was no compiler for Pascal, Wirth wrote one in Pascal, then "interpreted" it in his head and "ran" it on paper. A secretary might indeed have been handy.)
MUDs, wherefrom the MMORPGs truly spring, have had addictive substances for quite a while now.
I think BatMUD has had tobacco addiction for as long as I can remember. Which would be something like since 1994-6 or something. Quite likely also earlier than that.
I don't know, but Wario Ware, Inc. covers pretty many genres and most of those microgames require pretty simple controls (some may need A+control, but a pretty large number is winnable with either A presses or just the controller).
And when you've got Wario Ware and Tetris, you've got everything! =)
Errrr... is this astroturfing, or do real people actually use words like "leverage"?
Well, what every expert said was "content matters, not the gimmics", the folks said "Yeah! Let's create some content!" with same results as you imagine.
Content still matters. However, it should indeed be remembered that "content" isn't created, it's grown. This was the fundamental error for some people.
Individual pieces of work aren't "content" as such - just like an almost completely empty house with a single chair in it isn't exactly well-decorated. Content is the whole thing.
In other words, you can't win if you think that your purpose is to throw some shit in and then go shouting "Yes, we have Content". Everyone should still be focusing individual bits and pieces rather than the whole picture.
Bah! You have no idea of true scambaiting art. This is art. A gripping tale of dark dark horror and greed. Guaranteed to spook you silly. Or not. =)
Seriously, anyone can get scammers to pose for silly photographs, but this tale was something completely extraordinary.
Okay, the 419eater scams are still funny enough, though... I particularly liked the one where the scammer tried to pass off photo from Vatican as a photo of their church =)
So this pppppowerbook is worth two thousand dollars? And cheap materials are worth tens of thousands?
Somehow, I'm imagining that the folks who came up with this way to trick scammers are laughing now, but wait 'til the scammer sells the pppppppowerbook to some modern art gallery or computer history museum. The scammers sure are smart! They can easily make zillions!
Or they started coding it before they wrote the story in detail, or even came up with the names. Might not surprise me at all.
Completely untrue! Are you, perhaps, too too young to remember Taco Hell? =)
And ages ago, Slashdot was full of usual blog-type stuff like "Somerandomnickname writes, "Someone actually mentioned Linux somewhere". Wow." Or something along those lines. =)
Bloxs... Blosxom, despite its tricky name, can operate as a CGI script or it can generate static HTML and RSS files. A small and sweet app overall, nicely extendable too. All you need is Perl to run it locally.
Yup, Blosxom rules. I couldn't believe my eyes when I first saw it. Doesn't necessarily need a RDBMS (often these things only do MySQL! Horror!) to work, and can generate static pages, so it works on my web space provider (who also support CGI, but I preferred to do static pages).
Perhaps because the only good implementations out there right now are Vorbis reference implementations - especially on encoder side. It's interesting to see if anyone can come up with even better encoder. All that matters to me is that the files are compatible.
Never question the Japanese, for they shall forevermore be shrouded in mystery and strangeness, and the other world simply cannot understand them at all times. =) Not that it's needed. They have strange naming ideas but good tech and software, and that's what matters.
Dunno about breaking the Message-ID part. I replaced the "groups" with "groups-beta" and it still found the message just as well - looks like the links won't get broken that easily when they roll this out. I do hope they make it still easy to find messages by Message-ID, though...
It's still difficult.
Suppose I had a Linux C++ application that I wanted to port to Windows and have a nice native GUI for it. Decided to use MinGW for this. There, I have all of the necessary headers and stuff for the low-level work. I can compile the base stuff easily on Windows, or even cross-compile them on Linux - basically most parts of the code that don't need UI or the obscure OS features.
Now, the UI work. I can get WTL from sourceforge, but how the heck do I get all of the headers and libraries it depends on (ATL and the others)? How difficult it is to obtain them? How do I get them to work nicely with g++? What kind of restrictions are there to using these libraries? Etc etc etc...
At this point, I'd probably just ditch it and go for wxwindows or something =)
Actually, I'd very much like to know this - I've heard some of the libraries are available through Microsoft with some restrictions. Anyone know better?
Yeah, it skips sometimes; I just turn the buffering off! (Strange but true - when there's a lot of skipping, it seems to be *less* skippy with buffering on... Must be that the Area 51 technology from just a bit to the left from Rigel was cleverly disguised as cheap CD players and sent to the world with "Made in Taiwan" written on the back, or something. Nobody knew where to look from.)
As for jogging - music is overrated for jogging in my opinion. People are supposed to drop off all unnecessary junk and listen nothing but the sounds of their environment. The Nature. Or something like that. But yeah, if music is what you need for that, solid state memory rules. =)
God I guessed there would be a WarioWare title. What a better way to start the life of the "innovative two-display system" than to use it for something completely crazy? First, they laid down the ideas on how the dual-screen functionality could be used, and then, with WarioWare, they'll make a title that tosses the rules away for a while to do something unusual. Or at least that's what I'm definitely going to expect if it's going to be a genuine WarioWare title. I'm certainly hoping it will work that way =)