Actually, the way I read the "information broker" article doesn't seem like a problem. It's essentially an extension of the browser's current role in deciding to open a PDF in Acrobat Reader, Preview, or Evince, to open an email link in your default mail client, etc. Only now it'll be able to open based on data in a page. Got an ISBN number? Open it in your favorite online bookstore. Contact info? Add it to your address book.
It's a lot like Opera's ability to highlight text on a page and send it to a search engine, a dictionary, or a translation service.
As long as the sentiment behind "The personal toolbar is the personal toolbar, not the whorebar" remains intact, and as long as it's possible to choose competing different services/apps, the risk of corporate encroachment is minimal.
Does that include the ability to only run on Vista?
That's probably a safe bet. Windows XP would actually be out of "mainstream support" today (more than 5 years from release) if Vista hadn't been delayed. Microsoft decided a year ago to extend support for XP indefinitely. Now that Vista's out, WinXP is the new Win2k. It seem likely that XP will drop into "extended support" (i.e. security fixes only, and only for XP Pro) sometime during the "18-24 month" timeframe cited for IE8
And your bounce bounces into someone elses mailbox
No, the point is that it doesn't bounce. It's rejected in the SMTP transaction, which means that the connecting server just sees "500 user unknown" before it even transmits the message body -- no new bounce message is created. And if the message you're rejecting is a bounce in the first place, it should have a null sender, which means that the connecting server won't try to generate a new bounce either.
When you get a chance, check out the current nightly trunk builds. Just after Firefox 3 alpha 1, they merged in the "reflow branch" which includes a bunch of CSS improvements and passes Acid2.
You got half-way there with the part about microformats being created by others. The key is that microformats (the "extend" part in this case) discussed so far are described openly and free to use.
If Firefox starts supporting, say, hCard and hCalendar by making it possible to send the data to the Thunderbird address book or the calendar app of your choice, there's nothing to stop Opera, Apple, or indeed Microsoft from doing the same thing. Other browser developers don't have to reverse-engineer the features, or sign an NDA, or pay for a patent license.
Embrace is good. Extend is OK too, when done in a way that makes the third step, "Extinguish," difficult to do.
Assuming it's the "Places" concept they were working on for Firefox 2 and postponed, it boils down to this:
Put bookmarks (and history, etc.) in a lightweight database instead of a big long HTML file. This will make it possible for the user to store a lot more bookmarks before performance degrades, will make it easier to search, etc.
psychological and intelligent decision-making processes in the brain probably supercede animal instincts.
Actually, one of the hazards of sex is that animal instincts often do override intelligent decision-making processes. That's why many college campuses try to make condoms as available as possible (through conveniently-located vending machines, mainly, though student health at my college would give them out for free): They know students are going to be having sex, and when you're in bed with someone, you don't want to stop, get in the car, run to the corner store and buy a new pack of condoms.
Using your own address makes you more traceable and means you have to deal with bounces, complaints, etc.
Using a forged address saves you that inconvenience.
Completely bogus addresses will have a low throughput, because it's trivial for a receiving server to check whether a domain name exists or not.
Verifying a specific address at a real domain, however, is more involved.
Solution: Use a bogus address at a real domain name.
This solution expresses itself in both throwaway domains (where the spammer registers it for cheap, figuring they only need it for one spam run) and forged addresses using bystander's domains. Forging is cheaper, since you don't have to register a domain, and while it's illegal, enforcement is rare.
The problem of invalid bounces drops dramatically if you set up your incoming server so that invalid addressees are rejected with a "User unknown" note at SMTP time. If you're using Sendmail with a virtual user table, this is as easy as adding the following at the end of the file
@example.com error:nouser 550 5.1.1 User unknown
It's important to do this on the server that accepts mail from the outside. If you have a setup with an antispam/virus gateway that then relays to an internal server, you need to make the gateway aware of the valid/invalid addresses.
By rejecting invalid senders in the SMTP transaction, you only get bounces from the few messages that forged an actual sender. In my experience, the addresses tend to look like ashawuiefgfyig@example.com, so most of the bounces will just disappear into the ether(net).
As a die-hard gamer, I am ready to buy games on Linux and OSX, and NOT upgrade to Vista.
OK...but are the restrictions on Vista's LUAs any more difficult to work with than using root privileges to install a game on Linux, or entering an administrative password to allow installation on OSX?
There's also applications. I tend to use a Gnome desktop, because I find it easier to work with, but I prefer certain KDE apps like KMail and aKregator.
Fedora's media check has historically been pickier than it needs to be. If the media passes, you know it's good, but there are things that will make it fail other than an actual bad disc. They suggested booting the installer with "linux ide=nodma" if it fails with the default settings. The release notes for FC5 and FC6 don't include this note, so they may have fixed the bug. I can't say myself, since FC4 was the last time I installed from CD.
Could be. I was going from memory, and I ended up not actually burning any ISOs on any of the machines where I installed FC6. I did a net install on one, and a hard drive based install on the other two. Either way, the exact number didn't leave much of an impression.
I will admit to the "one CD" limit being wishful thinking on my part, assuming that this would start being feasible now that everything will be "one single repository." Besides, if they're talking about a "desktop spin" and a "KDE spin," that implies leaving KDE off one set of media, which implies paring down the contents of the install media.
Still, I should have used a phrase like "With any luck..." or "One hopes..." instead of "Presumably..."
There does seem to be a tendency among some Linux users to look at desktop OSes in terms of Windows and Linux, rather than Windows, Linux/*BSD, and Mac. I suspect it's a holdover from the days when most Linux users were expatriate Windows users, installing Linux on their PCs, and Mac was this other thing that ran on totally different hardware.
As for Safari and Linux, at least you can get a half-way decent approximation with Konqueror. It's far from perfect, of course, since (IIRC) WebKit and KHTML are being developed on two separate tracks that occasionally feed back into one another. Plus there's different fonts, functionality that's outside the rendering engine, and features that rely on Mac system libraries.
There are ways to force IE 5.5 and IE 6 to display a PNG through another image library which does handle alpha transparency correctly. The one I usually use is PNG Behavior, because it's unobtrusive -- the only change it requires is assigning a class to alpha PNGs and adding one proprietary CSS rule.
Basically, it replaces the image with a blank one, then loads a filter which displays the actual image in the background. Since the filter can display alpha transparency, you get an alpha-blended image. This works on native installations of IE 5.5 and IE6. On WINE, though, the filter doesn't work, so all it succeeds in is replacing the image with a blank.
(Oddly, I found the same thing happening to the stand-alone copies of IE on my Windows box when I upgraded from the IE7 release candidate to the final version. It prompted me to finally set up VirtualPC.)
I do most of my development on my personal web projects on my Linux box at home. Every once in a while I will fire up the Windows box and test things in IE. But I have a copy of IE6 installed through Crossover Office that I can use to verify that, say, a CSS change I've just made does what I think it does. The main problems I've encountered are fonts and the filter problem I mentioned here.
Basically, I use the WINE copy for (pun not intended) sanity checks, and a native copy for serious testing.
you will be able to develop and test webpages across almost all major browsers (IE 5-7, Firefox, Opera) on one Linux box!
If you do your main development on a Linux box, and want to test minor changes in IE as you make them (major changes and final testing should still be done on a native system if possible), it's a lot more convenient to fire up a copy of IE in WINE than to move over to another box or reboot into Windows.
One of the problems I've had running IE6 through WINE (not through ies4linux, just a stock Crossover install) is that the filter-based workarounds to trick IE5.5 and IE6 into displaying alpha-transparent PNG images correctly just doesn't work. IIRC, it's because those methods force IE to display the image through an ActiveX control which isn't present on Linux systems. It replaces the image with a blank one, but doesn't display the alpha-blended background.
This shouldn't be an issue with IE7, but it does make it difficult to test layouts that use alpha PNG and rely on the IE6 workaround.
It's good to know that they've got conditional comments working, though. That's always been the trick with running multiple IEs on Windows. You have to tweak the registry, or else each IE engine will parse them as if it were the most recent one installed on the system.
can the latter category necessarily afford a $3,000 expense every three years to buy a commodity PC with a newer Windows OS and the newer hardware it requires and a Mac with a newer Mac OS X and the newer hardware it requires?
If all you need to do is test websites, you don't need the latest and greatest hardware. Get a cheap $300 PC. It'll have the latest version of Windows. It won't run nifty games, but it'll handle IE7 fine. Buy a Mac Mini for $600. Get a KVM switch (a one-time expense) so you can share the keyboard, mouse and monitor. Now your expenses are more like $900 every three years.
And if you really don't want to buy new computers, there are sites like BrowserCam, , Browsershots, and iCapture that will at least test layout on other platforms.
Hey, that's nothing. In my day, I had to walk 14 miles through the snow, uphill, just to get to Slashdot. It's actually easier to get to now that it's gone downhill!
Slashdot Mirror
Actually, the way I read the "information broker" article doesn't seem like a problem. It's essentially an extension of the browser's current role in deciding to open a PDF in Acrobat Reader, Preview, or Evince, to open an email link in your default mail client, etc. Only now it'll be able to open based on data in a page. Got an ISBN number? Open it in your favorite online bookstore. Contact info? Add it to your address book.
It's a lot like Opera's ability to highlight text on a page and send it to a search engine, a dictionary, or a translation service.
As long as the sentiment behind "The personal toolbar is the personal toolbar, not the whorebar" remains intact, and as long as it's possible to choose competing different services/apps, the risk of corporate encroachment is minimal.
That's probably a safe bet. Windows XP would actually be out of "mainstream support" today (more than 5 years from release) if Vista hadn't been delayed. Microsoft decided a year ago to extend support for XP indefinitely. Now that Vista's out, WinXP is the new Win2k. It seem likely that XP will drop into "extended support" (i.e. security fixes only, and only for XP Pro) sometime during the "18-24 month" timeframe cited for IE8
No, the point is that it doesn't bounce. It's rejected in the SMTP transaction, which means that the connecting server just sees "500 user unknown" before it even transmits the message body -- no new bounce message is created. And if the message you're rejecting is a bounce in the first place, it should have a null sender, which means that the connecting server won't try to generate a new bounce either.
When you get a chance, check out the current nightly trunk builds. Just after Firefox 3 alpha 1, they merged in the "reflow branch" which includes a bunch of CSS improvements and passes Acid2.
You got half-way there with the part about microformats being created by others. The key is that microformats (the "extend" part in this case) discussed so far are described openly and free to use.
If Firefox starts supporting, say, hCard and hCalendar by making it possible to send the data to the Thunderbird address book or the calendar app of your choice, there's nothing to stop Opera, Apple, or indeed Microsoft from doing the same thing. Other browser developers don't have to reverse-engineer the features, or sign an NDA, or pay for a patent license.
Embrace is good. Extend is OK too, when done in a way that makes the third step, "Extinguish," difficult to do.
Assuming it's the "Places" concept they were working on for Firefox 2 and postponed, it boils down to this:
Put bookmarks (and history, etc.) in a lightweight database instead of a big long HTML file. This will make it possible for the user to store a lot more bookmarks before performance degrades, will make it easier to search, etc.
Are you using Safari 0.5 beta or something? I'm fairly certain Safari has had tabs since the beginning.
You must be new here.
I believe that study has been done already. I believe it was published by the journal, HotOrNot.com
Actually, one of the hazards of sex is that animal instincts often do override intelligent decision-making processes. That's why many college campuses try to make condoms as available as possible (through conveniently-located vending machines, mainly, though student health at my college would give them out for free): They know students are going to be having sex, and when you're in bed with someone, you don't want to stop, get in the car, run to the corner store and buy a new pack of condoms.
Or better yet, 6 pictures at 6 different points in their cycles. Fertility is a range of probabilities, not a boolean value.
There's also a mundane reason for it:
This solution expresses itself in both throwaway domains (where the spammer registers it for cheap, figuring they only need it for one spam run) and forged addresses using bystander's domains. Forging is cheaper, since you don't have to register a domain, and while it's illegal, enforcement is rare.
The problem of invalid bounces drops dramatically if you set up your incoming server so that invalid addressees are rejected with a "User unknown" note at SMTP time. If you're using Sendmail with a virtual user table, this is as easy as adding the following at the end of the file
@example.com error:nouser 550 5.1.1 User unknown
It's important to do this on the server that accepts mail from the outside. If you have a setup with an antispam/virus gateway that then relays to an internal server, you need to make the gateway aware of the valid/invalid addresses.
By rejecting invalid senders in the SMTP transaction, you only get bounces from the few messages that forged an actual sender. In my experience, the addresses tend to look like ashawuiefgfyig@example.com, so most of the bounces will just disappear into the ether(net).
OK...but are the restrictions on Vista's LUAs any more difficult to work with than using root privileges to install a game on Linux, or entering an administrative password to allow installation on OSX?
There's also applications. I tend to use a Gnome desktop, because I find it easier to work with, but I prefer certain KDE apps like KMail and aKregator.
Fedora's media check has historically been pickier than it needs to be. If the media passes, you know it's good, but there are things that will make it fail other than an actual bad disc. They suggested booting the installer with "linux ide=nodma" if it fails with the default settings. The release notes for FC5 and FC6 don't include this note, so they may have fixed the bug. I can't say myself, since FC4 was the last time I installed from CD.
Could be. I was going from memory, and I ended up not actually burning any ISOs on any of the machines where I installed FC6. I did a net install on one, and a hard drive based install on the other two. Either way, the exact number didn't leave much of an impression.
I will admit to the "one CD" limit being wishful thinking on my part, assuming that this would start being feasible now that everything will be "one single repository." Besides, if they're talking about a "desktop spin" and a "KDE spin," that implies leaving KDE off one set of media, which implies paring down the contents of the install media.
Still, I should have used a phrase like "With any luck..." or "One hopes..." instead of "Presumably..."
There does seem to be a tendency among some Linux users to look at desktop OSes in terms of Windows and Linux, rather than Windows, Linux/*BSD, and Mac. I suspect it's a holdover from the days when most Linux users were expatriate Windows users, installing Linux on their PCs, and Mac was this other thing that ran on totally different hardware.
As for Safari and Linux, at least you can get a half-way decent approximation with Konqueror. It's far from perfect, of course, since (IIRC) WebKit and KHTML are being developed on two separate tracks that occasionally feed back into one another. Plus there's different fonts, functionality that's outside the rendering engine, and features that rely on Mac system libraries.
There are ways to force IE 5.5 and IE 6 to display a PNG through another image library which does handle alpha transparency correctly. The one I usually use is PNG Behavior, because it's unobtrusive -- the only change it requires is assigning a class to alpha PNGs and adding one proprietary CSS rule.
Basically, it replaces the image with a blank one, then loads a filter which displays the actual image in the background. Since the filter can display alpha transparency, you get an alpha-blended image. This works on native installations of IE 5.5 and IE6. On WINE, though, the filter doesn't work, so all it succeeds in is replacing the image with a blank.
(Oddly, I found the same thing happening to the stand-alone copies of IE on my Windows box when I upgraded from the IE7 release candidate to the final version. It prompted me to finally set up VirtualPC.)
I do most of my development on my personal web projects on my Linux box at home. Every once in a while I will fire up the Windows box and test things in IE. But I have a copy of IE6 installed through Crossover Office that I can use to verify that, say, a CSS change I've just made does what I think it does. The main problems I've encountered are fonts and the filter problem I mentioned here.
Basically, I use the WINE copy for (pun not intended) sanity checks, and a native copy for serious testing.
From the article summary:
If you do your main development on a Linux box, and want to test minor changes in IE as you make them (major changes and final testing should still be done on a native system if possible), it's a lot more convenient to fire up a copy of IE in WINE than to move over to another box or reboot into Windows.
One of the problems I've had running IE6 through WINE (not through ies4linux, just a stock Crossover install) is that the filter-based workarounds to trick IE5.5 and IE6 into displaying alpha-transparent PNG images correctly just doesn't work. IIRC, it's because those methods force IE to display the image through an ActiveX control which isn't present on Linux systems. It replaces the image with a blank one, but doesn't display the alpha-blended background.
This shouldn't be an issue with IE7, but it does make it difficult to test layouts that use alpha PNG and rely on the IE6 workaround.
It's good to know that they've got conditional comments working, though. That's always been the trick with running multiple IEs on Windows. You have to tweak the registry, or else each IE engine will parse them as if it were the most recent one installed on the system.
If all you need to do is test websites, you don't need the latest and greatest hardware. Get a cheap $300 PC. It'll have the latest version of Windows. It won't run nifty games, but it'll handle IE7 fine. Buy a Mac Mini for $600. Get a KVM switch (a one-time expense) so you can share the keyboard, mouse and monitor. Now your expenses are more like $900 every three years.
And if you really don't want to buy new computers, there are sites like BrowserCam, , Browsershots, and iCapture that will at least test layout on other platforms.
Hey, that's nothing. In my day, I had to walk 14 miles through the snow, uphill, just to get to Slashdot. It's actually easier to get to now that it's gone downhill!