> Not just any modern OS, the BSDs, *nixes, and Windows > all have IPv6 support going back a decade. I'm not > sure about the classic Mac OS, though.
The last major-version release of "classic" MacOS (9) is over a decade old now, so if that's the timeframe you're working with you could probably consider just OS X.
I don't happen to know how well 10.0 supported IPv6, though.
> One in my town had its zoning amended to shut down the > drive thru [...] and went under 6 months or so later.
I'm a little surprised they didn't close it immediately (unless they were trying to appeal or something). McDonald's typically does 80% or more of their business in the drive-through (well, in North America; it might be different in some parts of the world). So except for fairly special cases (like, inside a mall), they generally don't locate where they don't expect to be able to keep a drive-through running.
> There are precious few alternatives. How else on the > net can you easily solicit contributions / donations?
Okay, I know this is going to sound a bit wacky and hopelessly out of the dernier cri web-2.0 fashion loop, but I'm going to go out on a limb here and suggest... perhaps you could just, umm, you know, ask people to send you money?
I know, it's crazy talk. Obviously, nobody would ever donate money to anything if they couldn't use PayPal to do it. Clearly, the whole economy would collapse without PayPal, because there's no other way to send money anywhere.
> was told that Pizza Hut didn't do much research > on site location, but simply put stores near McDonalds
I don't know if Pizza Hut really does that, but if they did, I bet it would work pretty well. McDonald's consistently picks good locations. Frequently they have the single perfectly optimal location in town (for towns that are too small to need more than one McDs).
There are five state routes that run through the city where I live (Galion). There's one McDonald's. You can see the McDonald's sign from all five state routes. They're also directly in front of the most popular grocery store's parking lot. (And yes, Pizza Hut is about a block and a half away, albeit around a corner.) Wendy's is downtown and gets significantly less traffic past their business (less local traffic, as well as less out-of-town traffic), not to mention less parking space. They get a bunch of extra business for Oktoberfest (which is held downtown), but that's only once a year. The rest of the time, the McDonald's location is clearly superior.
The next town up the road (Crestline) has two state routes that go through it, one north-south and one east-west. They run together for exactly one block. McDonald's is on that block. (There's no Pizza Hut, but I don't think there's one anywhere else in Crestline either.) It's *obviously* the best location in town.
I could go on, but you don't have to take my word for it. Start looking at where McDonald's is in every town near you, and see how many towns you have to look at before you find one where the McDonald's (if present) is in a mediocre location.
Have you ever seen a McDonald's closed for anything other than remodeling or expansion? I haven't. It's partly their advertising campaign (which is so extensive and so consistent and so effective, every marketing student should be required to study them for at least a full semester), but there are other factors too, and consistently picking good locations is one of them, I'm convinced.
> One day my dog can trade in the market, generate > enough income to pay for its dog food and let me retire.
Yes, but you have to know *which* day. There's only one day that produces the desired outcome. If you give your dog the money on any *other* day and let him play the market, he'll just blow all your money on worthless penny stocks and dying businesses.
> I know of 1 person with a facebook profile who isn't an attention whore,
I have a Facebook profile.
> and they use it just so they can spy on others.
In my case, I use it to read updates that a friend of mine is posting on there. He grew up here (Galion, Ohio) but is currently living in Vermont, attempting to plant a church there (yes, in Vermont, really; nobody ever accused him of avoiding difficult endeavors). He chose Facebook as the venue to post his updates. Personally, I would have preferred a site that lets people just stop by and read without signing up (say, Blogspot), but that's not what he selected. He's not a computer geek really, so Facebook is probably just what he happened to have heard of.
So yeah, I wanted to read his updates, and therefore I have a Facebook profile. I never post anything on it, though, and I hardly even bother looking at the various invitations and requests people send me. I just read Tom's home missions updates once a week or so, and that's basically it.
> > Then when random weird stuff started happening (e.g., all > > their icons disappeared), they didn't know why. > > I don't know why, either.
Nobody does. Some programmer on the Active Desktop project went to work drunk one day, would be my guess.
> What are you talking about?
Oh, I thought you knew, since you were talking about Active Desktop.
In Windows 98, if Active Desktop got turned on, it did arbitrary weird stuff from time to time, without provocation. Most of the misbehavior involved subverting the normal functionality of explorer.exe in some way, but since explorer.exe is responsible for such a huge percentage of the Windows UI, its being the culprit was not always immediately obvious. One day you'd turn the computer on, and the shortcut icons on your desktop would all be gone. No recycle bin, no My Computer, nothing. Another day you'd be using the computer and all of a sudden, right in the middle of whatever you were doing, the taskbar would just vanish. These are just examples. Whatever Active Desktop did, it was generally unexpected and unwanted, the kind of stuff that would make you think somebody was playing a prank on your computer. Once you turned Active Desktop off, everything would go back to normal. I saw this any number of times, on various computers, over the years Windows 98 was popular. Typically it happened when the user downloaded an image from the internet and didn't know they needed to convert it to a bmp before using it as wallpaper. Two or three days later, bang, weird stuff started happening.
As far as I know, this doesn't happen in Windows XP, though I've never left Active Desktop turned on long enough to be really sure about that. But we have a number of XP systems at work, so if it did happen on XP, I'd probably know by now.
> I continued using Netscape up until Firefox ("Phoenix" at > the time, beta), and have used it as my browser ever since.
I never really used Netscape 6 and 7 much. They lagged too far behind the unbranded Mozilla suite, not just in terms of features but also bug fixes. There was a general decline in crashiness from every few minutes circa 1998 down to virtually never, so that you can now leave a single browser window open for months at a time, by about 2006.
For a while around the turn of the century, the improvement in stability was so rapid that frequently even a nightly build would be noticeably more stable than the previous release. Meanwhile, Netscape's branded release was usually several releases behind, so it pretty much sucked pond water. IMO, this lag is what finally killed the Netscape brand (not that it was in great shape by then anyway).
The individual browser release that I used for the longest time, of course, was Navigator 4.08. I used that as my primary browser until the Mozilla suite became stable enough to actually use, sometime in 2000.
Wait, you used IE6, and you tried Firefox 1.0, and it didn't stun you with its relative awesomeness?
Wow. That's just... astonishing. Talk about an alien perspective!
I didn't switch over to Firefox as my primary browser until at least 1.5, but that's because I was using the old Mozilla suite, which I had been using since the late nineties. I certainly would have preferred Firefox 1.0 to IE6. Even the earliest alpha releases of Phoenix made IE6 look like a steaming heap of refuse. I can't imagine trying to use IE6 for daily browsing. Just using it for a few minutes on a new install until you could get another browser downloaded was painful. (Not quite as painful as earlier versions of IE, granted. I'm pretty sure IE4 was the worst browser ever.)
> You do realize that Firefox is ad-supported as well
Okay, yes, technically, but not in the horrible nagwareish way Opera used to be, with a big fat ad banner taking away 15% or so of the space that would otherwise be used for showing web pages. That was just terrible.
The way Firefox and modern versions of Opera handle this is unobtrusive and does not degrade the user's experience.
> Ability to natively open FTP sites as if they were folders
Yeah, Emacs had that in the eighties. By the time IE5 came out it could also transparently access remote directories and files via a number of other protocols, including ssh.
> and Active Desktop which nobody used
ITYM which nobody used *on purpose*.
People *frequently* got it turned on by mistake, often without even realizing it, typically because trying to use a JPEG or PNG image as wallpaper prompted you to turn it on and they just mindlessly clicked yes (as in, "yes, I want to put this picture on my wallpaper") without realizing the consequences. Then when random weird stuff started happening (e.g., all their icons disappeared), they didn't know why.
> There's nothing wrong with the OS providing a service > as universal as "viewing HTML content"
If it had only been plain old HTML content that it rendered (not also plugins and ActiveX and who only knows what else), it wouldn't have been as much of a problem.
Don't get me wrong: I'm very happy with IE8 as an upgrade from earlier versions of IE. As a webmaster, I am thrilled that I no longer have to maintain an entire extra stylesheet just to tell IE *not* to do a bunch of hopelessly stupid things that no other browser would ever contemplate doing. I consider IE8 to be a great success in that regard. And yeah, from a security standpoint, it's not perfect, but it's MUCH better than IE6.
Both the username and the domain are are going to have some impact, of course. Of the email addresses below, which ones make the most positive impressions?
Both the username and the domain matter. And no, the domain doesn't have to be a vanity domain. But a real ISP domain is generally better than a free-webmail domain, and a company or organization name is better yet, *especially* if it's relevant to the kind of work you're applying for.
And, of course, the email address matters more if you're applying for IT work than if you're applying for, say, assembly-line factory work.
No, what you do is, you get one of those extra-super-large jars of peanut butter, replace the lid with aluminum foil, attach a cheap analog alarm clock to the side with duct tape, wrap cheap jumper cables around the jar, pack it in a suitcase (fill up the remaining empty space with clothes), and take it with you on your next vacation. If they question you about why you have this stuff in your luggage, explain that some guy on the internet suggested it.
> Speed = Energy. In a turn and burn contest > the pilot with the must[ITYM most] energy wins
I'm not entirely convinced of that.
The information I have read on the subject suggests that other factors (maneuverability, armament) can be just as important as speed, if not more important. For example, the A-10 Thunderbolt II is said to be relatively good in a dogfight, notwithstanding the undeniable fact that it is slower than a typical fighter. It's slow, but it can turn on a dime (well enough to dodge heat-seeking missiles without the use of countermeasures), can continue flying after absorbing more machine-gun fire than is typical for a fighter, and carries fairly impressive weaponry, especially for its era.
Speed isn't everything. It can be important, but it isn't everything.
> Only programs I've ever had use MORE [RAM] > were Photoshop... and VMWare
Apparently you've not used Inkscape. Holy cow does that thing use a lot of memory. (It's not too bad otherwise, and I do have enough RAM that it's not a big problem. But it does use a lot, oh, yes.)
> > The UI will likely never be quite as fast due to XUL, > Therein lies the problem.
So go use Chrome. Or Lynx.
As for me, I like XUL. The kinds of extensions it makes possible supply a number of pretty major features, some of which I don't want to be without. I don't care if it makes certain things 20% slower. It still saves me time, overall, because it improves usability and workflows.
I don't care if the next version of Chrome loads every page in Planck time, I still don't want to use a browser that doesn't have Nuke Anything and Flash Block and Web Developer and Image Zoom and Rikaichan and so on.
Heck, bookmark keywords alone (a feature built into Gecko since the nineties) save me *WAY* more time than I lose to slow performance even when using SeaMonkey (which performs significantly worse than Firefox).
Real-world usability is more important than a technical performance advantage.
> Generally speaking, the only theories that have been proven > wrong are ones based on incorrect observations or when people > have simply made stuff of. (Which can hardly be called a > scientific theory.)
On the contrary, such ideas are called scientific theories all the time. Perhaps you haven't been paying attention to some of the pablum that's been put forward as "scientific" lately.
> Aristotle wasn't wrong
Of Aristotle's teachings, probably the most famous is that all matter is made up of four basic elements: earth, water, air, and fire. This is the main notion I was talking about, which the entire Western world held onto for centuries because of his authority. It's an idea that does not have any truth to it at all, period. Not that he didn't have other ideas as well, but this is the big one.
> what's wrong with the Control Panel interface that hinders developers
You mean besides the fact that it's been completely rearranged and the various bits renamed and specific settings moved from place to place so many times nobody can find anything?
It took me three minutes playing around in the Windows Seven control panel just to figure out how to change the TCP/IP settings. They're in a different place from Vista, where they were in a different place from XP, which in turn put them in a different place from both 98 and 2000. Almost every setting in the control panel has been moved around like that, and some of them have seen worse, being moved into places where you can only get to them through "Advanced Settings" menu items and/or links in sidebars that don't even show at all if you have Windows Explorer set to classic folders.
I can definitely see how the developers could become frustrated with it, especially if they were working on a not-yet-released future version at work while still using an actually released version elsewhere (say, at home, or on the laptop). The inconsistency could be maddening.
> DNS can be messed up and mail will still function. > Say you have a hosted domain but it lacks an mx > record. Mail will still go out. So the server on > the other end needs to make a choice. Throw it > away or pass it through.
It doesn't have to be a binary choice based on one criterion. You can use a number of different checks (does the envelope sender match the From field, does either of them match the HELO domain, does the HELO domain match the sending IP address, is the message text or HTML, does the sending domain provide SPF records and if so do they match, is there a valid subject line and if so does it match one of these regular expressions, have any of our users sent mail to this domain in the last N days,...) to drive a variable which, if it passes a certain threshhold, can trigger other effects (greylist, check against IP blacklists, run a virus scan, whatever). You can even have multiple threshholds: if the message fails 40-60% of the checks you might greylist it with a short delay, and if if fails 61-80% you might greylist it with a longer delay, but if it fails more than 80% of the checks you might reject it out of hand, or send a "please confirm" reply that requires user interaction, or even go into teargrube mode. And you might weight some of the checks more heavily than others. For example, I'd penalize HTML mail much more heavily than mail with mismatched From and envelope sender, and domains to which I've sent mail would get a pretty big break, and so on.
Spammers use multiple techniques. If we want to keep up in the arms race, we're going to have to use multiple techniques to fight back.
> Yeah I hate the way anybody can just walk > past my house and drop stuff in the letterbox.
That's not the problem. Indeed, that's an intentional and useful design feature.
The problem with SMTP is that it costs you more to maintain your mailbox than it costs the senders to keep dropping junk in it.
With a better design of mail protocol, advertisers would still be able to send you whatever junk they want, but it would cost them more to send it than it costs you to maintain a mailbox for receiving it. The obvious way to do this is to set up the protocol so that the sending mail server tells the recipient's mail server, "I have a message for [your address], with such-and-such a message ID." The recipient's mail server then keeps track of this information until the user checks their mail. The user's mailreader then gets this information, and it can either be set up to retrieve all the messages from their various sources and store them locally (offline mailreader), or else it can be set up to show the user a list, and individual messages are only retrieved from the sender when the user clicks on them (online reader). A hybrid setup would also be possible (e.g., if the sender is in my address book, go ahead and retrieve the message). This design makes the sender responsible to store the message indefinitely (although messages could have an expiration date after which they are no longer available), and the recipient's ISP does *not* have to store the whole message, just the metadata. This is a rough outline, of course, and there are a number of details that would have to be ironed out, but it doesn't matter, because it would never be adopted anyway, because it wouldn't be backward compatible with SMTP.
> There just isn't much room for energy storage inside a dragonfly.
There isn't all that much mass, either.
> They must have commendably efficient ways of staying in the air,
I'm sure the ratio of surface area to mass has something to do with it.
What I want to know is, when we say "Ocean-crossing" here, does he just mean they're flying from India to the Maldives (about 500 miles offshore)? Because the word "crossing" normally implies "from one side to the other", which, when it comes to oceans, would generally be a rather greater distance (multiple thousands of miles).
Slashdot Mirror
> Not just any modern OS, the BSDs, *nixes, and Windows
> all have IPv6 support going back a decade. I'm not
> sure about the classic Mac OS, though.
The last major-version release of "classic" MacOS (9) is over a decade old now, so if that's the timeframe you're working with you could probably consider just OS X.
I don't happen to know how well 10.0 supported IPv6, though.
> One in my town had its zoning amended to shut down the
> drive thru [...] and went under 6 months or so later.
I'm a little surprised they didn't close it immediately (unless they were trying to appeal or something). McDonald's typically does 80% or more of their business in the drive-through (well, in North America; it might be different in some parts of the world). So except for fairly special cases (like, inside a mall), they generally don't locate where they don't expect to be able to keep a drive-through running.
> There are precious few alternatives. How else on the
> net can you easily solicit contributions / donations?
Okay, I know this is going to sound a bit wacky and hopelessly out of the dernier cri web-2.0 fashion loop, but I'm going to go out on a limb here and suggest... perhaps you could just, umm, you know, ask people to send you money?
I know, it's crazy talk. Obviously, nobody would ever donate money to anything if they couldn't use PayPal to do it. Clearly, the whole economy would collapse without PayPal, because there's no other way to send money anywhere.
> was told that Pizza Hut didn't do much research
> on site location, but simply put stores near McDonalds
I don't know if Pizza Hut really does that, but if they did, I bet it would work pretty well. McDonald's consistently picks good locations. Frequently they have the single perfectly optimal location in town (for towns that are too small to need more than one McDs).
There are five state routes that run through the city where I live (Galion). There's one McDonald's. You can see the McDonald's sign from all five state routes. They're also directly in front of the most popular grocery store's parking lot. (And yes, Pizza Hut is about a block and a half away, albeit around a corner.) Wendy's is downtown and gets significantly less traffic past their business (less local traffic, as well as less out-of-town traffic), not to mention less parking space. They get a bunch of extra business for Oktoberfest (which is held downtown), but that's only once a year. The rest of the time, the McDonald's location is clearly superior.
The next town up the road (Crestline) has two state routes that go through it, one north-south and one east-west. They run together for exactly one block. McDonald's is on that block. (There's no Pizza Hut, but I don't think there's one anywhere else in Crestline either.) It's *obviously* the best location in town.
I could go on, but you don't have to take my word for it. Start looking at where McDonald's is in every town near you, and see how many towns you have to look at before you find one where the McDonald's (if present) is in a mediocre location.
Have you ever seen a McDonald's closed for anything other than remodeling or expansion? I haven't. It's partly their advertising campaign (which is so extensive and so consistent and so effective, every marketing student should be required to study them for at least a full semester), but there are other factors too, and consistently picking good locations is one of them, I'm convinced.
> One day my dog can trade in the market, generate
> enough income to pay for its dog food and let me retire.
Yes, but you have to know *which* day. There's only one day that produces the desired outcome. If you give your dog the money on any *other* day and let him play the market, he'll just blow all your money on worthless penny stocks and dying businesses.
> Were they high during this experiment?
Only if you count research funding as a drug. (Which, I suppose, is arguable.)
> I wonder, what it is now... "Angelina Jolie"? "Bruce Willis"?
Actually, it's Tokugawa Mujahibamidad Prolszinoczewski Cohen now. They wanted to be more culturally inclusive.
> I know of 1 person with a facebook profile who isn't an attention whore,
I have a Facebook profile.
> and they use it just so they can spy on others.
In my case, I use it to read updates that a friend of mine is posting on there. He grew up here (Galion, Ohio) but is currently living in Vermont, attempting to plant a church there (yes, in Vermont, really; nobody ever accused him of avoiding difficult endeavors). He chose Facebook as the venue to post his updates. Personally, I would have preferred a site that lets people just stop by and read without signing up (say, Blogspot), but that's not what he selected. He's not a computer geek really, so Facebook is probably just what he happened to have heard of.
So yeah, I wanted to read his updates, and therefore I have a Facebook profile. I never post anything on it, though, and I hardly even bother looking at the various invitations and requests people send me. I just read Tom's home missions updates once a week or so, and that's basically it.
> > Then when random weird stuff started happening (e.g., all
> > their icons disappeared), they didn't know why.
>
> I don't know why, either.
Nobody does. Some programmer on the Active Desktop project went to work drunk one day, would be my guess.
> What are you talking about?
Oh, I thought you knew, since you were talking about Active Desktop.
In Windows 98, if Active Desktop got turned on, it did arbitrary weird stuff from time to time, without provocation. Most of the misbehavior involved subverting the normal functionality of explorer.exe in some way, but since explorer.exe is responsible for such a huge percentage of the Windows UI, its being the culprit was not always immediately obvious. One day you'd turn the computer on, and the shortcut icons on your desktop would all be gone. No recycle bin, no My Computer, nothing. Another day you'd be using the computer and all of a sudden, right in the middle of whatever you were doing, the taskbar would just vanish. These are just examples. Whatever Active Desktop did, it was generally unexpected and unwanted, the kind of stuff that would make you think somebody was playing a prank on your computer. Once you turned Active Desktop off, everything would go back to normal. I saw this any number of times, on various computers, over the years Windows 98 was popular. Typically it happened when the user downloaded an image from the internet and didn't know they needed to convert it to a bmp before using it as wallpaper. Two or three days later, bang, weird stuff started happening.
As far as I know, this doesn't happen in Windows XP, though I've never left Active Desktop turned on long enough to be really sure about that. But we have a number of XP systems at work, so if it did happen on XP, I'd probably know by now.
> I continued using Netscape up until Firefox ("Phoenix" at
> the time, beta), and have used it as my browser ever since.
I never really used Netscape 6 and 7 much. They lagged too far behind the unbranded Mozilla suite, not just in terms of features but also bug fixes. There was a general decline in crashiness from every few minutes circa 1998 down to virtually never, so that you can now leave a single browser window open for months at a time, by about 2006.
For a while around the turn of the century, the improvement in stability was so rapid that frequently even a nightly build would be noticeably more stable than the previous release. Meanwhile, Netscape's branded release was usually several releases behind, so it pretty much sucked pond water. IMO, this lag is what finally killed the Netscape brand (not that it was in great shape by then anyway).
The individual browser release that I used for the longest time, of course, was Navigator 4.08. I used that as my primary browser until the Mozilla suite became stable enough to actually use, sometime in 2000.
Wait, you used IE6, and you tried Firefox 1.0, and it didn't stun you with its relative awesomeness?
Wow. That's just... astonishing. Talk about an alien perspective!
I didn't switch over to Firefox as my primary browser until at least 1.5, but that's because I was using the old Mozilla suite, which I had been using since the late nineties. I certainly would have preferred Firefox 1.0 to IE6. Even the earliest alpha releases of Phoenix made IE6 look like a steaming heap of refuse. I can't imagine trying to use IE6 for daily browsing. Just using it for a few minutes on a new install until you could get another browser downloaded was painful. (Not quite as painful as earlier versions of IE, granted. I'm pretty sure IE4 was the worst browser ever.)
> You do realize that Firefox is ad-supported as well
Okay, yes, technically, but not in the horrible nagwareish way Opera used to be, with a big fat ad banner taking away 15% or so of the space that would otherwise be used for showing web pages. That was just terrible.
The way Firefox and modern versions of Opera handle this is unobtrusive and does not degrade the user's experience.
> Ability to natively open FTP sites as if they were folders
Yeah, Emacs had that in the eighties. By the time IE5 came out it could also transparently access remote directories and files via a number of other protocols, including ssh.
> and Active Desktop which nobody used
ITYM which nobody used *on purpose*.
People *frequently* got it turned on by mistake, often without even realizing it, typically because trying to use a JPEG or PNG image as wallpaper prompted you to turn it on and they just mindlessly clicked yes (as in, "yes, I want to put this picture on my wallpaper") without realizing the consequences. Then when random weird stuff started happening (e.g., all their icons disappeared), they didn't know why.
> There's nothing wrong with the OS providing a service
> as universal as "viewing HTML content"
If it had only been plain old HTML content that it rendered (not also plugins and ActiveX and who only knows what else), it wouldn't have been as much of a problem.
Don't get me wrong: I'm very happy with IE8 as an upgrade from earlier versions of IE. As a webmaster, I am thrilled that I no longer have to maintain an entire extra stylesheet just to tell IE *not* to do a bunch of hopelessly stupid things that no other browser would ever contemplate doing. I consider IE8 to be a great success in that regard. And yeah, from a security standpoint, it's not perfect, but it's MUCH better than IE6.
Both the username and the domain are are going to have some impact, of course. Of the email addresses below, which ones make the most positive impressions?
lickme@demon.co.uk
jon_robertson@demon.co.uk
robert.grunwald@juno.com
robert.grunwald@donottaunthappyfunball.com
robert.grunwald@pobox.com
robert.grunwald@mozilla.org
americanidol418709874@yahoo.com
natalie_branson_jones@yahoo.com
americanidol418709874@gmail.com
natalie_branson_jones@gmail.com
americanidol418709874@jonesvilleconsulting.com
natalie@jonesvilleconsulting.com
Both the username and the domain matter. And no, the domain doesn't have to be a vanity domain. But a real ISP domain is generally better than a free-webmail domain, and a company or organization name is better yet, *especially* if it's relevant to the kind of work you're applying for.
And, of course, the email address matters more if you're applying for IT work than if you're applying for, say, assembly-line factory work.
No, what you do is, you get one of those extra-super-large jars of peanut butter, replace the lid with aluminum foil, attach a cheap analog alarm clock to the side with duct tape, wrap cheap jumper cables around the jar, pack it in a suitcase (fill up the remaining empty space with clothes), and take it with you on your next vacation. If they question you about why you have this stuff in your luggage, explain that some guy on the internet suggested it.
> Speed = Energy. In a turn and burn contest
> the pilot with the must[ITYM most] energy wins
I'm not entirely convinced of that.
The information I have read on the subject suggests that other factors (maneuverability, armament) can be just as important as speed, if not more important. For example, the A-10 Thunderbolt II is said to be relatively good in a dogfight, notwithstanding the undeniable fact that it is slower than a typical fighter. It's slow, but it can turn on a dime (well enough to dodge heat-seeking missiles without the use of countermeasures), can continue flying after absorbing more machine-gun fire than is typical for a fighter, and carries fairly impressive weaponry, especially for its era.
Speed isn't everything. It can be important, but it isn't everything.
> Only programs I've ever had use MORE [RAM] ... and VMWare
> were Photoshop
Apparently you've not used Inkscape. Holy cow does that thing use a lot of memory. (It's not too bad otherwise, and I do have enough RAM that it's not a big problem. But it does use a lot, oh, yes.)
> > The UI will likely never be quite as fast due to XUL,
> Therein lies the problem.
So go use Chrome. Or Lynx.
As for me, I like XUL. The kinds of extensions it makes possible supply a number of pretty major features, some of which I don't want to be without. I don't care if it makes certain things 20% slower. It still saves me time, overall, because it improves usability and workflows.
I don't care if the next version of Chrome loads every page in Planck time, I still don't want to use a browser that doesn't have Nuke Anything and Flash Block and Web Developer and Image Zoom and Rikaichan and so on.
Heck, bookmark keywords alone (a feature built into Gecko since the nineties) save me *WAY* more time than I lose to slow performance even when using SeaMonkey (which performs significantly worse than Firefox).
Real-world usability is more important than a technical performance advantage.
Twelve years ago, I would have said an AOL email address would be just about the worst thing you could put on a resume.
But you know, today there's hotmail, and people put that on resumes all the time. Somehow, AOL doesn't seem so bad any more.
> Generally speaking, the only theories that have been proven
> wrong are ones based on incorrect observations or when people
> have simply made stuff of. (Which can hardly be called a
> scientific theory.)
On the contrary, such ideas are called scientific theories all the time. Perhaps you haven't been paying attention to some of the pablum that's been put forward as "scientific" lately.
> Aristotle wasn't wrong
Of Aristotle's teachings, probably the most famous is that all matter is made up of four basic elements: earth, water, air, and fire. This is the main notion I was talking about, which the entire Western world held onto for centuries because of his authority. It's an idea that does not have any truth to it at all, period. Not that he didn't have other ideas as well, but this is the big one.
> When you say "more widely used", you mean
> as a percentage of published code?
No, I mean in terms of the total number of people using it.
> what's wrong with the Control Panel interface that hinders developers
You mean besides the fact that it's been completely rearranged and the various bits renamed and specific settings moved from place to place so many times nobody can find anything?
It took me three minutes playing around in the Windows Seven control panel just to figure out how to change the TCP/IP settings. They're in a different place from Vista, where they were in a different place from XP, which in turn put them in a different place from both 98 and 2000. Almost every setting in the control panel has been moved around like that, and some of them have seen worse, being moved into places where you can only get to them through "Advanced Settings" menu items and/or links in sidebars that don't even show at all if you have Windows Explorer set to classic folders.
I can definitely see how the developers could become frustrated with it, especially if they were working on a not-yet-released future version at work while still using an actually released version elsewhere (say, at home, or on the laptop). The inconsistency could be maddening.
> DNS can be messed up and mail will still function.
...) to drive a variable which, if it passes a certain threshhold, can trigger other effects (greylist, check against IP blacklists, run a virus scan, whatever). You can even have multiple threshholds: if the message fails 40-60% of the checks you might greylist it with a short delay, and if if fails 61-80% you might greylist it with a longer delay, but if it fails more than 80% of the checks you might reject it out of hand, or send a "please confirm" reply that requires user interaction, or even go into teargrube mode. And you might weight some of the checks more heavily than others. For example, I'd penalize HTML mail much more heavily than mail with mismatched From and envelope sender, and domains to which I've sent mail would get a pretty big break, and so on.
> Say you have a hosted domain but it lacks an mx
> record. Mail will still go out. So the server on
> the other end needs to make a choice. Throw it
> away or pass it through.
It doesn't have to be a binary choice based on one criterion. You can use a number of different checks (does the envelope sender match the From field, does either of them match the HELO domain, does the HELO domain match the sending IP address, is the message text or HTML, does the sending domain provide SPF records and if so do they match, is there a valid subject line and if so does it match one of these regular expressions, have any of our users sent mail to this domain in the last N days,
Spammers use multiple techniques. If we want to keep up in the arms race, we're going to have to use multiple techniques to fight back.
> Yeah I hate the way anybody can just walk
> past my house and drop stuff in the letterbox.
That's not the problem. Indeed, that's an intentional and useful design feature.
The problem with SMTP is that it costs you more to maintain your mailbox than it costs the senders to keep dropping junk in it.
With a better design of mail protocol, advertisers would still be able to send you whatever junk they want, but it would cost them more to send it than it costs you to maintain a mailbox for receiving it. The obvious way to do this is to set up the protocol so that the sending mail server tells the recipient's mail server, "I have a message for [your address], with such-and-such a message ID." The recipient's mail server then keeps track of this information until the user checks their mail. The user's mailreader then gets this information, and it can either be set up to retrieve all the messages from their various sources and store them locally (offline mailreader), or else it can be set up to show the user a list, and individual messages are only retrieved from the sender when the user clicks on them (online reader). A hybrid setup would also be possible (e.g., if the sender is in my address book, go ahead and retrieve the message). This design makes the sender responsible to store the message indefinitely (although messages could have an expiration date after which they are no longer available), and the recipient's ISP does *not* have to store the whole message, just the metadata. This is a rough outline, of course, and there are a number of details that would have to be ironed out, but it doesn't matter, because it would never be adopted anyway, because it wouldn't be backward compatible with SMTP.
So yeah, SMTP is broken, but it can't be fixed.
> There just isn't much room for energy storage inside a dragonfly.
There isn't all that much mass, either.
> They must have commendably efficient ways of staying in the air,
I'm sure the ratio of surface area to mass has something to do with it.
What I want to know is, when we say "Ocean-crossing" here, does he just mean they're flying from India to the Maldives (about 500 miles offshore)? Because the word "crossing" normally implies "from one side to the other", which, when it comes to oceans, would generally be a rather greater distance (multiple thousands of miles).