Call bullshit all you want, doesn't change the facts. That may not have been the reason the mail was getting flagged, it was just my guess. All I know is that when I sent the mail (php app using the regular php mail() command and the -f switch to force a particular return address) to the students and bcc'ed even just 5 people during the development phase, the messages got flagged by Hotmail's spam filter. As soon as I changed to doing a loop and making referencing one recipient at a time in the to: field, it got through.
If you've got a better explanation, I'd like to hear it.
here is no way they'd know you were BCC'ing anyone, because (by definition), BCCs aren't seen by the recipient (either the user or mailserver.)
Umm, the fact that the bcc'd recipient doesn't appear in the too field in the headers? As you point out, bcc's aren't seen by the recipient, and that includes recipients that are included as part of the bcc. In other words, if I bcc an e-mail to myself, my address will not show up in the to: line. Pretty simple check as a matter of fact. For example, here are the headers of a test message I sent to my hotmail account as the to: recipient and to my normal personal account as a bcc: recipient:
Return-Path: Received: from [12.217.253.155] (account )
by admin.mwci.net (CommuniGate Pro WebUser 4.1)
with HTTP id 4417436; Sun, 17 Aug 2003 14:38:54 -0700 From: Subject: bcc test To: jd142@hotmail.com X-Mailer: CommuniGate Pro WebUser Interface v.4.1 Date: Sun, 17 Aug 2003 14:38:54 -0700 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="ISO-8859-1"; format="flowed" Content-Transfer-Encoding: 8bit
As you can see, my normal account does not show up in the to: field because it was bcc'd. It is trivial to write a program to check to see if the recipient e-mail address is in the to: line of the headers. If it is not (and if it is not a part of the cc field), then it is reasonable to assume that the recipient was in the bcc field. Duh.
So I suggest that if you are going to make shit up about email, you learn to view some headers, learn a little bit about what the mail clients see as part of the headers, see what's being transferred and think for a few seconds before you start flaming someone.
Getting around spam filters is not just trying to get your e-mail client in an inbox that really matches one of the filters you have personally made.
Here's a real world example. I wrote an application so that staff in our college could go to a web page and send mail to the students of our college, either all students or by class year. Not wanting every person to see every other person's e-mail, I initially set this program up to bcc everyone and send a copy to the Deans as the to: recipients so they would know what the students got and I put a generic address as the from: so the students could hit reply and have it go to a central account but they could also see the deans' addresses to e-mail them.
Unfortunately, this got flagged by places like Hotmail and Yahoo as spam because I had just bcc'ed a large number of people.
So I had to send the messages out one at a time as individual messages, not as one message with a huge number of recipients.
I believe it is this kind of spam filter, cases where there is a legitimate reason to send mail to thousands of recipients without letting the recipients see each other's addresses, that the original poster was referring to.
Actually, people are more confused by something that looks very similar to what they are used to but has minor differences and less confused by something that is completely different.
It's a game of expectations. If something looks exactly like Windows, then I expect to be able to do everything exactly the same. For example, setting the wallpaper. In KDE this is under Look and Feel, but in Windows it's in Display. So a windows user is frustrated because things aren't in the same place even though everything else looks just like Windows.
If Gnome looks sufficiently different, then people don't have expectations of how things work, where common utilities are located, etc. So the user will not be as frustrated because his or her pre-existing knowledge isn't wrong.
Just so you know, you can set up your own windows update server pretty easily. It's hard to set up a server that will work with the red carpet service; at least that's what I've been told and what I've read. And then you simply use Group Policies to tell all the computers on your network to use your windows update server for their updates. You can control whether or not they reboot after a patch, whether the reboot is forced, exactly which updates you want to push down to your users, and whether or not the users have any choice in the matter. The update server will apply patches in the right order and get dependencies as needed.
It depends what you mean by more than once. As I understand it, and I've seen this happen, if the first program the user runs is WordPerfect, it uses MFC32.dll from c:\windows\system\mfc32.dll. Let's say that this is version 3.4.2 of mfc32.dll That dll is then loaded in to memory. Next, the user runs foo.exe. c:\program files\foo has a version 3.4.5 of mfc32.dll in it just in case. Since mfc32.dll is already loaded into memory, foo.exe will end up using that version, 3.4.2. I've actually seen minor differences in version numbers of dll's cause problems. So Eudora would work, but only if you loaded it before WordPerfect; if you loaded WP first, then Eudora would have an error. So to the extent that only one version of the dll gets loaded at a time, then you can't use more than one. But if I ran foo.exe first, it would find mfc32.dll in its directory and load 3.4.5 into memory.
At least that's my understanding of how it works; I'm not claiming I know everything by a long shot. I have seen instances where you had to load programs in a certain order though because of minor differences in the dll versions. I suspect there's a way to write your program so as to require it to load the dll that you specify instead of depending on windows, but I have no clue how to do that.
In theory, this will change in windows, so that things are more like they were in the dos days -- stick an app in a directory and put everything it needs there. You end up with multiple copies of the same file, but the apps are compartmentalized nicely. And the installs should be as easy as the are on a Mac; simply copy the files into the programs directory and you're off.
What about dynamically linking, but with fall through.
Here's (basically) how it works in the windows world. You start an application that needs a function from a specific dll. First, the application looks to see if the dll is loaded into memory. If the dll is not loaded into memory, then the application looks in the directory it is being run from, i.e., c:\program files\foo\foo.exe would look for bar.dll in c:\program files\foo. If not found there, it would look in the windows sytem directories, %windir%\system32 and %windir%\system. Well written programs also keep track of their use of a dll that's store in the system32 directory. There's a section of the registry that tracks the number of programs that use a dll. So if foo.exe uses mfc32.dll, during the installation it increments the counter in the registry. When a program being uninstalled sees that there are no other programs using a shared dll, it asks the user if the dll should be deleted.
I simply don't know how it works in Linuxland. But it would be nice something similar could be done. Each application ships with all of the libraries it needs. If, during the install process, it sees that the libraries are already present, then it doesn't need to install them(it just tells some central location that it needs a library) or it could let the user choose to install them in the same directory as the app.
One of the big drawbacks to the linux way of doing things that I've seen is that applications tend to be installed into the same directories, everything in/usr/bin,/bin,/usr/sbin, etc. I know that this is supposed to neatly separate applications based on whether they are a part of the base system, a user add-on, etc., but it seems that putting each application in its own directory makes more sense. I also understand that part of the reason for this is that many people use the command line and don't want to type/usr/bin/mozilla-1.4/mozilla just to start an app. But Linux on the desktop will have shortcuts that have the correct path. And I know that you can put symlinks into a directory in your path. So you could put/symlinks in your path and have symlinks in there to all the applications you use.
In this situation, you'd be better off exporting the data and re-importing it into something useful, probably using perl (or whatever your preference is) to massage the data.
Then, once the data is in a more portable format, make a bootable windows 95 cd with all of the applications you need on it. I've done this before; it isn't the easiest thing to do, but it can be done. The hardest part is making a ram drive and assigning the temp folder to the ram drive so there's a writable temp directory for apps that require write access somewhere. This may be hard for a database program if it needs to write a lock file somewhere. Now you've got a cd that will boot on just about any computer. Keep one old pc around just in case and make multiple backups of the cd, store 2 in different offsite locations.
Alternately, use wine/vmware/dosemu/whatever to emulate windows 95 on linux.
Lack of easy support for my intellimouse explorer. I'm so used to using the side buttons to go forward and back when browsing the web it's jarring when I move from mozilla on windows to mozilla on linux.
Cool. Repeat my post 6 minutes later, mine was at 10:21 and Shaklee39's was at 10:27, and get up to a 4 mod. That's a hoot. If I took Slashdot seriously, I'd be upset.;)
What could the RIAA do about it? A lot since you are specifically sharing copyrighted material, and if you make the material available to a guest account so that anyone can see it, there's no real distinction between doing that and doing Kazaalite.
If you want to do something like that, there's a better way.
Set up an ftp server or some other share. Leave one directory visible to anonymous. Encrypt the username and password as part of that directory. For example, ftp://my.warez.net/idrt-qsdd
Would mean that the user name is user and the password is pass. Now, IANAL, but it seems to me that if you specifically deny the RIAA and any of its delegates from decrypting that, then if they figured out my supersecret encryption formula, they would be guilty of violating the DMCA. Simply put a license agreement in the public area that says that by accepting the license the user agrees that s/he will not reveal either the contents of the directory or the encryption method to the riaa, mpaa, their delegates, etc.
correlation != causation. Maybe there are a lot of people saying that AOL users say "me too" a lot. Had this discussion taken place on usenet, we would have incremented the count by two, yet neither one of us are aol users.;)
Google already does this to a certain degree, although I don't know if their Activity ranking takes into account replies to topics or just number of messages or what.
If you look at the Google Groups listings you'll see a rough measure of their activity as shown by a green bar. For example, if you look at the rec.arts.comics.* hierarchy you'll see rac.xbooks has no activity. And sure enough, if you go to that group you'll see 2 posts from 2003, 8 from 2002, and a handful of older ones. rac.european has an almost full bar and looking there shows 5-10 posts each month. The others have completely full bars showing lots of posts each day.
Maybe Google should explain better how the Activity rating works; I didn't see a mention in the faq. Or perhaps show more detail than just the green bar.
Part of the problem with the article is that it forgets that people need to learn how to use a search engine.
Looking up just "apple" is a bad idea no matter what you use. MSN isn't much better, but it does one thing right: it realizes there are different types of apples and offers links at the top to computers, fruits, quicktime and a vacation company.
It's always a good idea to be more specific about what you want: "apple orchard", "apple computers", "apple movies", etc.
The real problem is that coders that develop for IE rarely check how pages work in anything else
Which I've always found to be a bizarre way of doing things. I've found that ever since Mozilla.9 or thereabouts, that by coding to Mozilla, you will get a page that will work in pretty much every major browser. Granted, I'm not doing anything that fancy, just dynamic pages built from php/mysql that use javascript to manipulate the dynamic elements based on user choices. Since the DOM (at least the elements I use) is the same for mozilla, ie, konqueror (which means it should work for Macs now) and Opera, I don't have to worry about building browser detection in the scripts.
Because Mozilla is stricter about coding, you'll get better written code. IE let's the developer be sloppy, which produces sloppy pages. Mozilla is more strict, which forces me to produce better code.
I would assume that just like Mozilla let's you block graphics it will eventually let you turn off svg's. I also just found out about the flash blocker, Flash Cick to View. It's part of the Firebird extensions but also works great on plain mozilla 1.4 if you get it from the author's page.
With no popups, no ads and no flash, the web is usable.
True. But the people who use the guns can be held liable.
As an earlier poster pointed out, the problem with this is that a user's home computer could be providing kiddie porn. It's one thing to steal songs and software, but it's another thing to host pictures of some 7 year old getting raped. I don't want to even have the possibility of that happening, so I think I'll stick with another distributed client.
Legally, would host computers be analogous to the phone company -- a common carrier? If you use a telephone to plot to kill the president, the feds don't bust the phone company as part of the conspiracy. Just like they don't bust AOL for providing chat rooms for 35 year olds to pick up 12 year old English girls. Are people hosting files or parts of files like the phone company in the eyes of the law?
Right, but is there a difference between what they are doing: Here is the EULA [their offer]. Click ok to accept the offer[acceptance].
And what I am doing: Here is my counter offer. Notify my within 180 days if you do not accept through the postage paid envelope I have provided. (Realized I needed to add that last part.)
The big difference is that even in the first scenario, there is a small affirmitive act on the part of the end user to accept the license. My guess, but I don't know, is that you can't make a contract that says doing nothing signifies acceptance, even if there has been plenty of notice of the fact and an easy way to signify the decline.
However, I'd also bet there are acceptions to the rule. For example, instances of contracts that automatically renew or are automatically extended with the same or similar terms probably don't require an affirmitive action to accept the offer.
But I don't know the law. It's an interesting question. Some companies and governments are allowed to renegotiate their contracts with Microsoft. Why can't I? A partial answer is that it would take too many resources on their part for a little bit of gain. But, I respond, that's why I include the postage paid return envelope. They don't have to renegotiate, they just have to decline with the prepaid mailer I provide, so its very minimal on their part.
I've never tested it, but there's a Baysian spam filter plugin available for Squirrelmail.
Personally, I prefer systems that either put spam into one mailbox (like Mozilla Mail or Thunderbird) or that mark the message as spam in some way so that I can filter it (like adding [possible spam] to the message subject). The idea of a system just arbitrarily bouncing or deleting my mail on its own is troublesome. Right now, out of every 100 messages Thunderbird classifies as spam, 1 is a real message. If the messages were automatically deleted, I'd miss real mail.
No, to a degree the top poster is right. What we're talking about here is called an adhesion contract. IANAL, but usually such contracts are interpreted to the benefit of the person who cannot bargain. But it's been awhile, so someone may want to point out the nuances.
I've often wondered what would happen if you printed out an EULA, made changes to it, then sent it back to a company with the stipulation that if they do not contact you within 180 days that they signify their acceptance of your changes. You'd have to send it registered mail so you have a delivery receipt of course. And you wouldn't want to just send it to "Legal Department" at Microsoft. It would be an interesting experiment.
What would be the difference between what you did and what they did?
We often hear that software piracy costs X amount of dollars or music piract costs Y dollars every year. How are those numbers arrived at?
For example, let's say Pete Pirate grabs a copy of Microsoft Office and the latest Brittany Spears album from some site.
Microsoft has a list price of $499 for a full version of Microsoft Office XP Professional. But no one in the history of computing has ever paid full retail price for any copy of office. warehouse.com, about as mainstream as you can get sells it for $65 dollars less. You can get it for much less if you shop around or purchase it as part of an OEM deal.
Similarly, Brittany Spears's latest cd has a list price of $19. No one pays that either.
So did Pete pirate $520 worth of product or $447? That's about a 14% increase in damages.
And what's to stop a company from artificially inflating the MSRP of a product in order to inflate the damages?
I work in tech support at a college, and this is right on the money. If it isn't a chat program or word, they can't use it. I've sent students selfextracting zip files and they didn't know how to save an attachment in their webmail, as an example that happened just today.
What's worse is that they think they know what they're doing. 5 years ago, they were willing to admit they didn't know how to use a computer, while now they think that because they use hotmail, they're computer savvy.
Slashdot Mirror
Call bullshit all you want, doesn't change the facts. That may not have been the reason the mail was getting flagged, it was just my guess. All I know is that when I sent the mail (php app using the regular php mail() command and the -f switch to force a particular return address) to the students and bcc'ed even just 5 people during the development phase, the messages got flagged by Hotmail's spam filter. As soon as I changed to doing a loop and making referencing one recipient at a time in the to: field, it got through.
If you've got a better explanation, I'd like to hear it.
here is no way they'd know you were BCC'ing anyone, because (by definition), BCCs aren't seen by the recipient (either the user or mailserver.)
Umm, the fact that the bcc'd recipient doesn't appear in the too field in the headers? As you point out, bcc's aren't seen by the recipient, and that includes recipients that are included as part of the bcc. In other words, if I bcc an e-mail to myself, my address will not show up in the to: line. Pretty simple check as a matter of fact. For example, here are the headers of a test message I sent to my hotmail account as the to: recipient and to my normal personal account as a bcc: recipient:
Return-Path:
Received: from [12.217.253.155] (account )
by admin.mwci.net (CommuniGate Pro WebUser 4.1)
with HTTP id 4417436; Sun, 17 Aug 2003 14:38:54 -0700
From:
Subject: bcc test
To: jd142@hotmail.com
X-Mailer: CommuniGate Pro WebUser Interface v.4.1
Date: Sun, 17 Aug 2003 14:38:54 -0700
Message-ID:
MIME-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"; format="flowed"
Content-Transfer-Encoding: 8bit
As you can see, my normal account does not show up in the to: field because it was bcc'd. It is trivial to write a program to check to see if the recipient e-mail address is in the to: line of the headers. If it is not (and if it is not a part of the cc field), then it is reasonable to assume that the recipient was in the bcc field. Duh.
So I suggest that if you are going to make shit up about email, you learn to view some headers, learn a little bit about what the mail clients see as part of the headers, see what's being transferred and think for a few seconds before you start flaming someone.
Getting around spam filters is not just trying to get your e-mail client in an inbox that really matches one of the filters you have personally made.
Here's a real world example. I wrote an application so that staff in our college could go to a web page and send mail to the students of our college, either all students or by class year. Not wanting every person to see every other person's e-mail, I initially set this program up to bcc everyone and send a copy to the Deans as the to: recipients so they would know what the students got and I put a generic address as the from: so the students could hit reply and have it go to a central account but they could also see the deans' addresses to e-mail them.
Unfortunately, this got flagged by places like Hotmail and Yahoo as spam because I had just bcc'ed a large number of people.
So I had to send the messages out one at a time as individual messages, not as one message with a huge number of recipients.
I believe it is this kind of spam filter, cases where there is a legitimate reason to send mail to thousands of recipients without letting the recipients see each other's addresses, that the original poster was referring to.
Actually, people are more confused by something that looks very similar to what they are used to but has minor differences and less confused by something that is completely different.
It's a game of expectations. If something looks exactly like Windows, then I expect to be able to do everything exactly the same. For example, setting the wallpaper. In KDE this is under Look and Feel, but in Windows it's in Display. So a windows user is frustrated because things aren't in the same place even though everything else looks just like Windows.
If Gnome looks sufficiently different, then people don't have expectations of how things work, where common utilities are located, etc. So the user will not be as frustrated because his or her pre-existing knowledge isn't wrong.
Just so you know, you can set up your own windows update server pretty easily. It's hard to set up a server that will work with the red carpet service; at least that's what I've been told and what I've read. And then you simply use Group Policies to tell all the computers on your network to use your windows update server for their updates. You can control whether or not they reboot after a patch, whether the reboot is forced, exactly which updates you want to push down to your users, and whether or not the users have any choice in the matter. The update server will apply patches in the right order and get dependencies as needed.
Yep, that worked. I'm surprised Mandrake didn't install imwheel by default.
Thanks
Ok, so it has already changed. I thought it was going to be the next version of windows.
It depends what you mean by more than once. As I understand it, and I've seen this happen, if the first program the user runs is WordPerfect, it uses MFC32.dll from c:\windows\system\mfc32.dll. Let's say that this is version 3.4.2 of mfc32.dll That dll is then loaded in to memory. Next, the user runs foo.exe. c:\program files\foo has a version 3.4.5 of mfc32.dll in it just in case. Since mfc32.dll is already loaded into memory, foo.exe will end up using that version, 3.4.2. I've actually seen minor differences in version numbers of dll's cause problems. So Eudora would work, but only if you loaded it before WordPerfect; if you loaded WP first, then Eudora would have an error. So to the extent that only one version of the dll gets loaded at a time, then you can't use more than one. But if I ran foo.exe first, it would find mfc32.dll in its directory and load 3.4.5 into memory.
At least that's my understanding of how it works; I'm not claiming I know everything by a long shot. I have seen instances where you had to load programs in a certain order though because of minor differences in the dll versions. I suspect there's a way to write your program so as to require it to load the dll that you specify instead of depending on windows, but I have no clue how to do that.
In theory, this will change in windows, so that things are more like they were in the dos days -- stick an app in a directory and put everything it needs there. You end up with multiple copies of the same file, but the apps are compartmentalized nicely. And the installs should be as easy as the are on a Mac; simply copy the files into the programs directory and you're off.
What about dynamically linking, but with fall through.
/usr/bin, /bin, /usr/sbin, etc. I know that this is supposed to neatly separate applications based on whether they are a part of the base system, a user add-on, etc., but it seems that putting each application in its own directory makes more sense. I also understand that part of the reason for this is that many people use the command line and don't want to type /usr/bin/mozilla-1.4/mozilla just to start an app. But Linux on the desktop will have shortcuts that have the correct path. And I know that you can put symlinks into a directory in your path. So you could put /symlinks in your path and have symlinks in there to all the applications you use.
Here's (basically) how it works in the windows world. You start an application that needs a function from a specific dll. First, the application looks to see if the dll is loaded into memory. If the dll is not loaded into memory, then the application looks in the directory it is being run from, i.e., c:\program files\foo\foo.exe would look for bar.dll in c:\program files\foo. If not found there, it would look in the windows sytem directories, %windir%\system32 and %windir%\system. Well written programs also keep track of their use of a dll that's store in the system32 directory. There's a section of the registry that tracks the number of programs that use a dll. So if foo.exe uses mfc32.dll, during the installation it increments the counter in the registry. When a program being uninstalled sees that there are no other programs using a shared dll, it asks the user if the dll should be deleted.
I simply don't know how it works in Linuxland. But it would be nice something similar could be done. Each application ships with all of the libraries it needs. If, during the install process, it sees that the libraries are already present, then it doesn't need to install them(it just tells some central location that it needs a library) or it could let the user choose to install them in the same directory as the app.
One of the big drawbacks to the linux way of doing things that I've seen is that applications tend to be installed into the same directories, everything in
In this situation, you'd be better off exporting the data and re-importing it into something useful, probably using perl (or whatever your preference is) to massage the data.
;)
Then, once the data is in a more portable format, make a bootable windows 95 cd with all of the applications you need on it. I've done this before; it isn't the easiest thing to do, but it can be done. The hardest part is making a ram drive and assigning the temp folder to the ram drive so there's a writable temp directory for apps that require write access somewhere. This may be hard for a database program if it needs to write a lock file somewhere. Now you've got a cd that will boot on just about any computer. Keep one old pc around just in case and make multiple backups of the cd, store 2 in different offsite locations.
Alternately, use wine/vmware/dosemu/whatever to emulate windows 95 on linux.
Still a chance for a tech to make a killing.
Lack of easy support for my intellimouse explorer. I'm so used to using the side buttons to go forward and back when browsing the web it's jarring when I move from mozilla on windows to mozilla on linux.
Cool. Repeat my post 6 minutes later, mine was at 10:21 and Shaklee39's was at 10:27, and get up to a 4 mod. That's a hoot. If I took Slashdot seriously, I'd be upset. ;)
What could the RIAA do about it? A lot since you are specifically sharing copyrighted material, and if you make the material available to a guest account so that anyone can see it, there's no real distinction between doing that and doing Kazaalite.
;)
If you want to do something like that, there's a better way.
Set up an ftp server or some other share. Leave one directory visible to anonymous. Encrypt the username and password as part of that directory. For example, ftp://my.warez.net/idrt-qsdd
Would mean that the user name is user and the password is pass. Now, IANAL, but it seems to me that if you specifically deny the RIAA and any of its delegates from decrypting that, then if they figured out my supersecret encryption formula, they would be guilty of violating the DMCA. Simply put a license agreement in the public area that says that by accepting the license the user agrees that s/he will not reveal either the contents of the directory or the encryption method to the riaa, mpaa, their delegates, etc.
Who can be the first to crack my secret code?
correlation != causation. Maybe there are a lot of people saying that AOL users say "me too" a lot. Had this discussion taken place on usenet, we would have incremented the count by two, yet neither one of us are aol users.
Google already does this to a certain degree, although I don't know if their Activity ranking takes into account replies to topics or just number of messages or what.
If you look at the Google Groups listings you'll see a rough measure of their activity as shown by a green bar. For example, if you look at the rec.arts.comics.* hierarchy you'll see rac.xbooks has no activity. And sure enough, if you go to that group you'll see 2 posts from 2003, 8 from 2002, and a handful of older ones. rac.european has an almost full bar and looking there shows 5-10 posts each month. The others have completely full bars showing lots of posts each day.
Maybe Google should explain better how the Activity rating works; I didn't see a mention in the faq. Or perhaps show more detail than just the green bar.
There's also a toolbar extension for mozilla called PrefBar that is pretty similar.
Part of the problem with the article is that it forgets that people need to learn how to use a search engine.
Looking up just "apple" is a bad idea no matter what you use. MSN isn't much better, but it does one thing right: it realizes there are different types of apples and offers links at the top to computers, fruits, quicktime and a vacation company.
It's always a good idea to be more specific about what you want: "apple orchard", "apple computers", "apple movies", etc.
The real problem is that coders that develop for IE rarely check how pages work in anything else
.9 or thereabouts, that by coding to Mozilla, you will get a page that will work in pretty much every major browser. Granted, I'm not doing anything that fancy, just dynamic pages built from php/mysql that use javascript to manipulate the dynamic elements based on user choices. Since the DOM (at least the elements I use) is the same for mozilla, ie, konqueror (which means it should work for Macs now) and Opera, I don't have to worry about building browser detection in the scripts.
Which I've always found to be a bizarre way of doing things. I've found that ever since Mozilla
Because Mozilla is stricter about coding, you'll get better written code. IE let's the developer be sloppy, which produces sloppy pages. Mozilla is more strict, which forces me to produce better code.
I would assume that just like Mozilla let's you block graphics it will eventually let you turn off svg's. I also just found out about the flash blocker,
Flash Cick to View. It's part of the Firebird extensions but also works great on plain mozilla 1.4 if you get it from the author's page.
With no popups, no ads and no flash, the web is usable.
True. But the people who use the guns can be held liable.
As an earlier poster pointed out, the problem with this is that a user's home computer could be providing kiddie porn. It's one thing to steal songs and software, but it's another thing to host pictures of some 7 year old getting raped. I don't want to even have the possibility of that happening, so I think I'll stick with another distributed client.
Legally, would host computers be analogous to the phone company -- a common carrier? If you use a telephone to plot to kill the president, the feds don't bust the phone company as part of the conspiracy. Just like they don't bust AOL for providing chat rooms for 35 year olds to pick up 12 year old English girls. Are people hosting files or parts of files like the phone company in the eyes of the law?
I wish I had mod points. I've been waiting for a flash killer and I've been looking for some good tutorials.
Right, but is there a difference between what they are doing: Here is the EULA [their offer]. Click ok to accept the offer[acceptance].
And what I am doing: Here is my counter offer. Notify my within 180 days if you do not accept through the postage paid envelope I have provided. (Realized I needed to add that last part.)
The big difference is that even in the first scenario, there is a small affirmitive act on the part of the end user to accept the license. My guess, but I don't know, is that you can't make a contract that says doing nothing signifies acceptance, even if there has been plenty of notice of the fact and an easy way to signify the decline.
However, I'd also bet there are acceptions to the rule. For example, instances of contracts that automatically renew or are automatically extended with the same or similar terms probably don't require an affirmitive action to accept the offer.
But I don't know the law. It's an interesting question. Some companies and governments are allowed to renegotiate their contracts with Microsoft. Why can't I? A partial answer is that it would take too many resources on their part for a little bit of gain. But, I respond, that's why I include the postage paid return envelope. They don't have to renegotiate, they just have to decline with the prepaid mailer I provide, so its very minimal on their part.
I've never tested it, but there's a Baysian spam filter plugin available for Squirrelmail.
Personally, I prefer systems that either put spam into one mailbox (like Mozilla Mail or Thunderbird) or that mark the message as spam in some way so that I can filter it (like adding [possible spam] to the message subject). The idea of a system just arbitrarily bouncing or deleting my mail on its own is troublesome. Right now, out of every 100 messages Thunderbird classifies as spam, 1 is a real message. If the messages were automatically deleted, I'd miss real mail.
No, to a degree the top poster is right. What we're talking about here is called an adhesion contract. IANAL, but usually such contracts are interpreted to the benefit of the person who cannot bargain. But it's been awhile, so someone may want to point out the nuances.
I've often wondered what would happen if you printed out an EULA, made changes to it, then sent it back to a company with the stipulation that if they do not contact you within 180 days that they signify their acceptance of your changes. You'd have to send it registered mail so you have a delivery receipt of course. And you wouldn't want to just send it to "Legal Department" at Microsoft. It would be an interesting experiment.
What would be the difference between what you did and what they did?
We often hear that software piracy costs X amount of dollars or music piract costs Y dollars every year. How are those numbers arrived at?
For example, let's say Pete Pirate grabs a copy of Microsoft Office and the latest Brittany Spears album from some site.
Microsoft has a list price of $499 for a full version of Microsoft Office XP Professional. But no one in the history of computing has ever paid full retail price for any copy of office. warehouse.com, about as mainstream as you can get sells it for $65 dollars less. You can get it for much less if you shop around or purchase it as part of an OEM deal.
Similarly, Brittany Spears's latest cd has a list price of $19. No one pays that either.
So did Pete pirate $520 worth of product or $447? That's about a 14% increase in damages.
And what's to stop a company from artificially inflating the MSRP of a product in order to inflate the damages?
I work in tech support at a college, and this is right on the money. If it isn't a chat program or word, they can't use it. I've sent students selfextracting zip files and they didn't know how to save an attachment in their webmail, as an example that happened just today.
What's worse is that they think they know what they're doing. 5 years ago, they were willing to admit they didn't know how to use a computer, while now they think that because they use hotmail, they're computer savvy.