The parent was referring to the likes of So.Big which infected Windows boxes to setup SMTP zombies for the Spammers to use. I'm unaware of any *n*x vulnerability which has been exploited to create a network and spamboxes.
I suppose since I didn't qualify what kind of users jumped ship to KDE (non-lay-users).
GNOME was subjected to usability testing and new design goals were established. The [lay-]users who tested it wanted different default settings and fewer configuration dialogs. So GNOME 2.0 delivered on that. People who like simplified desktop environments (like those offered by Windows and Mac) liked GNOME 2.0+ more than GNOME 1.4-. People who don't care about newbi usability prefer GNOME 1.4- and KDE to GNOME 2.0+, Windows and Mac. The catalyst for the change was usability tests administered with [lay-]users.
If you look at reviews of KDE 3.2b, you'll see that the Conrtol Center is also starting to simplify itself a little more. You'll notice people saying 'finally this will make it easier to use for newbies, but it still has too many config options.' Of course you'll also notice people saying 'screw the UI experts, I want my options. You'd think newbies would want more options, otherwise they wouldn't be switching.' Ignoring the fact that it's the newbies themselves who said 'there are too many options, i can't figure out what they do.'
While I see what you're trying to say, a usability expert is the kind of person employed by a lab who pays hundreds of lay persons to use a system and say "this button doesn't work right" or "this feature should be more like this." The test administrators then take that feedback and analyse it with the gurus to produce the UI that reflects the feedback. This is what MS has done many times, and what Sun did with Gnome that led to the big philosophical change from 1.4 -> 2.0 that caused people to jump over to KDE.
I'm sure she's a perfectly fine lady, but is she also a GUI usability guru?
Being an expert would actually disqualify her as a good person for final approval. The guru is the one who will successfully implement the UI specs set forth by the lay-user.
If that involves getting Leslie involved in promoting OO.o I'm jumping ship.
PS - If you don't live in Austin, and/or don't know who Leslie is, you probably don't want to. He's no goatse guy guy in that he seems ok when described online. But trust me, barring great physical injury, nothing kills your buzz faster when stumbling down 6th St. to see him in stilleto heels and a g-string.
Considering all the shit movies that you can rent from Blockbuster now, they obviously aren't doing this anymore.... On a side note, I am only mad at them for only carrying R rates movies. Edited versions of Requium for a Dream and Showgirls doesn't make much sense. They're like walmart in that way, with [CLEAN] versions of rap CD's. LOL
That's exactly what he was talking about. The versions distributed to the video stores are edited to increase distribution. You are renting watered down versions.
Region codes are not a problem -- I have two DVD Players (cost $50 each). One is set for region 1 and one is set for region 2. Region codes were a problem when the DVD players were expensive.
That's silly. The only difference in those two players is a few bits to store what region they play. They don't decode different video formats or anything fancy. It's an arbitrary distinction that does nothing but cost you money. You wasted $50.
I think I get what you're saying, but you keep coming across as defending opt-in as an anti-spam solution by way of seeing how many people the email was sent to. They're unrelated, which you know.
So far the responses have been that we can't use this to block spam, but that was never the point; the point was to make spam illegal without affecting legitimate email.
The point of what? Opt-in systems or the new law? This thread is about opt-in systems so I'm going to apply it to that. The point of opt-in systems is the block spam. The point of the new law is to make spam illegal. The opt-in system throws the baby out with the bath water depending on the implementation. If the recipient has to mark valid senders, then he/she must either know the sender's address ahead of time or must approve them as mail is received. If the receiver hasn't approved a legitimate sender yet, than mail to the sender is rejected (thrown out with the bath water). If the receiver is using an on-demand system where approval is granted upon receipt of the mail, then the mail is quarantined where the receiver must look at it (and might throw it out with the bath water, as I have done using Yahoo's bulk mail system).
All I'm saying is that, LEGALLY, required opt-in for messages meeting the above condition (going to more than X people) wouldn't affect anybody who isn't spamming.
legally yes, but shakamojo presented it as, I think most of the people responding to him and you think, a technical solution. Shimmer disagreed. You responded with a post that seemed to say it was a good technical solution, because you can just filter based on the number of recipients.
The fact that you can disguise the number of messages sent out by sending them one at time doesn't change the total number.
that's true, but it has no relevance to an opt-in system, which filters on who sent it, not who receives it.
So to summarize, shakamojo said opt-in good, shimmer said opt-in bad, you said use number of recipients in opt-in, we said that won't work, you said no it won't but opt-in ok legally not technically. Is that right? If so then everyone except shakamojo agrees that opt-in is not a good long-term technical solution to spam, as it means going through the list of possible spam to approve senders or just loosing potentially legitimate mail because they weren't approved before they sent it.
So it's one at a time; the same email is still going out to millions of addresses. Therefore, it is spam.
how do the final SMTP server and/or the client filter know it's being sent 30 million times though? The headers on the mail you received have one address in the 'To:' header: yours. The SMTP server would have to start keeping checksums of message bodies to see how many copies of it are received in a given timeframe. That's a crappy solution, and is just as easy to workaround as sedning 30 million copies to individual addresses. (You just put the address in the body using a template when sending, and the body checksum will be different). It would just keep escalating. A better solution is to add a mechanism to protect against forged headers. Like a DNS record that receiving SMTP servers can use to verify the IP the email was sent from is a valid IP for mail from the domain in the headers.
If someone sent 30 million people the same or similar email then that is spam
I don't remember the last time I received spam with more than 1 recipient. So how would my filter (be it at the client or server level) know the difference based on that?
I suppose there are other programs that would rely on the windows api for html presentation, but could one port gecko so it takes care of this?
Nope, because the APIs were not published, which is a big part of what landed them in court. That's also the key difference that Windows apologists miss when they point out how many different browsers are included with a Linux distribution, or that Mac OS X ships with a browser. The browsers aren't truly interchangeable on Windows.
No they weren't. The "applications:" menu editing in Nautilus was disabled because it didn't work. But it was pretty simple to edit the menu with a text editor in/usr/share/applications.
scoff about RedHat's first suggestions like Gaim, OO.o, and other rubbish. Idiots!
Well that's a matter of opinion. Something has to be the default application. Red Hat decided to make a consistent desktop not only in terms of look and feel but in default applications. And they took hell for it. The default applications selected for both environments were the GNOME ones, obviously becase GNOME is the default DE.
And KDE or Gnome used to do this on RedHat 8 and/or 9 at least. Where you had your utils/apps/whatever in the main K/foot menu, PLUS another entry called "extras", which then cascaded into more utils/apps/whatever.
the infamous Extras menu of RH 8. It was static, so applications were either in their logical program group if they were default RH apps (or if you put them there intentionally), or they were in their logical program group in Extras. but it did suck. In RH 9 they fanned it out to having a sub menu on each program group. So you look at the Internet program group in the menu, and have "More Internet Apps", "Web Browser", "Email", and "Instant Messaging". So to find Internet applications you went Applications->Internet[->More]. Still a static system, apps didn't migrate in and out of More unless you explicitly moved them. Reduced clutter while retaining a logical layout.
considering that it(ie) can be pulled out without it affecting any major programs(beyond which use it directly) it's not that tightly integrated at all in windows.
No it can't. You can get the big blue 'e' off the desktop, but the rendering engine, and other components are contained in essential OS DLLs. The file browser is actually just another shell on top of the core IE. If you were to truly remove all IE code, you would have a non-funcitonal Windows system.
As someone on dot.kde.org pointed out (and I fully agree with) the ability to customize thing SHOULD not be messed with, because otherwise you go the GNOME/Windows way. KDE can act like almost any other DE if people want it to, and set it to do so.
The configuration options don't need to be removed, just slightly hideen. You could have each Control Center applet have an advanced tab with 50K extra options that 90% of the users don't mess with. Then the options are still there, very accessible, and out of the way so as not to intimidate the new users. And it does intimidate the new users. I'm a seasoned user and I find it's mostly just clutter. They could even have that first-run config wizard ask 'would you like to see advanced options in Control Center applets?' and leave them all mixed in. One of the things I think Red Hat did that made a large step in useability was adding the "More Applications" menu on each menu group. You see a small set of the defaults, and have like 5 alternatives in the submenu.
In order to maintain the one thing the use as a competitive edge (the integration of everything) without giving the same edge to competitors would be to make a totally proprietary system forked from FreeBSD. Of course, it would remove many of the features that make it secure, otherwise the integration wouldn't work. So why bother?
Unlike Windows, Mac OS X requires an administrator password to change certain configurations, run the system updater, and when installing new software. From a security perspective, this is another example of how Apple takes a proactive approach to system-level security. If a virus, remote hacker, or co-worker tries to install or reconfigure something on the system, theyre stymied without knowing the administrators password stored in the hardened System Keychain. (
Incidentally, this password is not the same as the Unix 'root' account password of the system's FreeBSD foundation, something that further enhances security.) In some ways, this can be seen as Mac OS X protecting a careless user from themself as well as others.
Not to say the original article was right (or wrong, never fully read it or the rebuttal) but it's shortsighted to assume criticism comes from zealous hate.
It's not too much of an assumption. The author of the orinigal piece said he was glad that there was finally a big vulnerability for Mac OS, and that he was tired of Mac users looking smug when SAMS edition Conquer the Internet in 12 Hours outlook viruses pass them over. The whole piece just had a tone of "I'm really sick of people bragging about Mac OS."
Because Hitler killed millions in the '30s and '40s, and nobody did anything to stop him, we should therefore do nothing to prevent the massacres occurring in Nigeria and elsewhere at the present moment?
Slashdot Mirror
The parent was referring to the likes of So.Big which infected Windows boxes to setup SMTP zombies for the Spammers to use. I'm unaware of any *n*x vulnerability which has been exploited to create a network and spamboxes.
double plus unable
Are the soul brother APIs just too beaucoup?
I suppose since I didn't qualify what kind of users jumped ship to KDE (non-lay-users).
GNOME was subjected to usability testing and new design goals were established. The [lay-]users who tested it wanted different default settings and fewer configuration dialogs. So GNOME 2.0 delivered on that. People who like simplified desktop environments (like those offered by Windows and Mac) liked GNOME 2.0+ more than GNOME 1.4-. People who don't care about newbi usability prefer GNOME 1.4- and KDE to GNOME 2.0+, Windows and Mac. The catalyst for the change was usability tests administered with [lay-]users.
If you look at reviews of KDE 3.2b, you'll see that the Conrtol Center is also starting to simplify itself a little more. You'll notice people saying 'finally this will make it easier to use for newbies, but it still has too many config options.' Of course you'll also notice people saying 'screw the UI experts, I want my options. You'd think newbies would want more options, otherwise they wouldn't be switching.' Ignoring the fact that it's the newbies themselves who said 'there are too many options, i can't figure out what they do.'
While I see what you're trying to say, a usability expert is the kind of person employed by a lab who pays hundreds of lay persons to use a system and say "this button doesn't work right" or "this feature should be more like this." The test administrators then take that feedback and analyse it with the gurus to produce the UI that reflects the feedback. This is what MS has done many times, and what Sun did with Gnome that led to the big philosophical change from 1.4 -> 2.0 that caused people to jump over to KDE.
Being an expert would actually disqualify her as a good person for final approval. The guru is the one who will successfully implement the UI specs set forth by the lay-user.
If that involves getting Leslie involved in promoting OO.o I'm jumping ship.
PS - If you don't live in Austin, and/or don't know who Leslie is, you probably don't want to. He's no goatse guy guy in that he seems ok when described online. But trust me, barring great physical injury, nothing kills your buzz faster when stumbling down 6th St. to see him in stilleto heels and a g-string.
That's exactly what he was talking about. The versions distributed to the video stores are edited to increase distribution. You are renting watered down versions.
That's silly. The only difference in those two players is a few bits to store what region they play. They don't decode different video formats or anything fancy. It's an arbitrary distinction that does nothing but cost you money. You wasted $50.
so wait, Debian isn't LSB compliant? What's with the LSB 'task' in tasksel then? (not trolling, just curious)
So far the responses have been that we can't use this to block spam, but that was never the point; the point was to make spam illegal without affecting legitimate email.
The point of what? Opt-in systems or the new law? This thread is about opt-in systems so I'm going to apply it to that. The point of opt-in systems is the block spam. The point of the new law is to make spam illegal. The opt-in system throws the baby out with the bath water depending on the implementation. If the recipient has to mark valid senders, then he/she must either know the sender's address ahead of time or must approve them as mail is received. If the receiver hasn't approved a legitimate sender yet, than mail to the sender is rejected (thrown out with the bath water). If the receiver is using an on-demand system where approval is granted upon receipt of the mail, then the mail is quarantined where the receiver must look at it (and might throw it out with the bath water, as I have done using Yahoo's bulk mail system).
All I'm saying is that, LEGALLY, required opt-in for messages meeting the above condition (going to more than X people) wouldn't affect anybody who isn't spamming.
legally yes, but shakamojo presented it as, I think most of the people responding to him and you think, a technical solution. Shimmer disagreed. You responded with a post that seemed to say it was a good technical solution, because you can just filter based on the number of recipients.
The fact that you can disguise the number of messages sent out by sending them one at time doesn't change the total number.
that's true, but it has no relevance to an opt-in system, which filters on who sent it, not who receives it.
So to summarize, shakamojo said opt-in good, shimmer said opt-in bad, you said use number of recipients in opt-in, we said that won't work, you said no it won't but opt-in ok legally not technically. Is that right? If so then everyone except shakamojo agrees that opt-in is not a good long-term technical solution to spam, as it means going through the list of possible spam to approve senders or just loosing potentially legitimate mail because they weren't approved before they sent it.
how do the final SMTP server and/or the client filter know it's being sent 30 million times though? The headers on the mail you received have one address in the 'To:' header: yours. The SMTP server would have to start keeping checksums of message bodies to see how many copies of it are received in a given timeframe. That's a crappy solution, and is just as easy to workaround as sedning 30 million copies to individual addresses. (You just put the address in the body using a template when sending, and the body checksum will be different). It would just keep escalating. A better solution is to add a mechanism to protect against forged headers. Like a DNS record that receiving SMTP servers can use to verify the IP the email was sent from is a valid IP for mail from the domain in the headers.
I don't remember the last time I received spam with more than 1 recipient. So how would my filter (be it at the client or server level) know the difference based on that?
Nope, because the APIs were not published, which is a big part of what landed them in court. That's also the key difference that Windows apologists miss when they point out how many different browsers are included with a Linux distribution, or that Mac OS X ships with a browser. The browsers aren't truly interchangeable on Windows.
No they weren't. The "applications:" menu editing in Nautilus was disabled because it didn't work. But it was pretty simple to edit the menu with a text editor in
scoff about RedHat's first suggestions like Gaim, OO.o, and other rubbish. Idiots!
Well that's a matter of opinion. Something has to be the default application. Red Hat decided to make a consistent desktop not only in terms of look and feel but in default applications. And they took hell for it. The default applications selected for both environments were the GNOME ones, obviously becase GNOME is the default DE.
the infamous Extras menu of RH 8. It was static, so applications were either in their logical program group if they were default RH apps (or if you put them there intentionally), or they were in their logical program group in Extras. but it did suck. In RH 9 they fanned it out to having a sub menu on each program group. So you look at the Internet program group in the menu, and have "More Internet Apps", "Web Browser", "Email", and "Instant Messaging". So to find Internet applications you went Applications->Internet[->More]. Still a static system, apps didn't migrate in and out of More unless you explicitly moved them. Reduced clutter while retaining a logical layout.
No it can't. You can get the big blue 'e' off the desktop, but the rendering engine, and other components are contained in essential OS DLLs. The file browser is actually just another shell on top of the core IE. If you were to truly remove all IE code, you would have a non-funcitonal Windows system.
in most other worlds, it means fucking ugly.
The configuration options don't need to be removed, just slightly hideen. You could have each Control Center applet have an advanced tab with 50K extra options that 90% of the users don't mess with. Then the options are still there, very accessible, and out of the way so as not to intimidate the new users. And it does intimidate the new users. I'm a seasoned user and I find it's mostly just clutter. They could even have that first-run config wizard ask 'would you like to see advanced options in Control Center applets?' and leave them all mixed in. One of the things I think Red Hat did that made a large step in useability was adding the "More Applications" menu on each menu group. You see a small set of the defaults, and have like 5 alternatives in the submenu.
In order to maintain the one thing the use as a competitive edge (the integration of everything) without giving the same edge to competitors would be to make a totally proprietary system forked from FreeBSD. Of course, it would remove many of the features that make it secure, otherwise the integration wouldn't work. So why bother?
emphasis added.
It's not too much of an assumption. The author of the orinigal piece said he was glad that there was finally a big vulnerability for Mac OS, and that he was tired of Mac users looking smug when SAMS edition Conquer the Internet in 12 Hours outlook viruses pass them over. The whole piece just had a tone of "I'm really sick of people bragging about Mac OS."
Without commenting on the validity any opnions regarding the justification of the war, that is the best rebuttal of google evidence I've ever seen.
turn them into giant fan blades for cooling the new uber-l337 opteron system.
Godwin's Law?