I thought I'd take the opportunity to mention the following idea here: Some levels of security are definitely worth having in just about any context and some are just a hassle in some environments. Since the issue of having to log in even if your intended system role is an Insecure Workstation/Desktop PC was brought up... To Linux, users are just numbers, with 0 having absolute privilege and others basically having their own stuff... there really isn't anything beyond convention stating that every user must have their own, secure account... In fact, it is perfectly possible and permissible to create an account "guest", with no password, and sufficient privileges to do the various basic activities that the end-user needs to perform.
That being said, it's not much of a stretch to suggest a login tool (for X; those who prefer text environments probably won't mind simply typing guest to start) which can be configured to automatically log in as guest, or to provide a button to do so directly at the login screen. The "guest" account would be disallowed for telnet, ftp, etc. and any security risks would probably be considerably fewer than those present in Windows 9x environments. Actually, I'm a little surprised a program like this hasn't been made yet. (Maybe it has... anyone know of an implementation of something like this?)
In my opinion, user-friendliness issues for Linux generally stem from what hasn't been done yet, not what can't be.
I'm not certain it's really fair to say that there is no legal protection from spam simply because one can't legally say who is allowed to read public documents. Telemarketing and related issues (e.g. UCE) can be considered a form of harrassment. One is free to speak, but that does not mean one can force people to listen to oneself. There are already legal restrictions on telemarketing; there's a good site for info on this, but I've forgotten the address...
Well, all I can say is... You don't have to give them any info. If you tell them you don't want to give them that information, they can (and as a matter of customer service do) continue the transaction without it. I used to work there back at the end of 1997 so this came up a few times.
I believe the currently applicable law you want is the DMCA, which has been discussed on Slashdot (check Microsoft's request to Slashdot regarding the Kerberos-related posts). I haven't finished reading it but I think it required certain specific things either to be retained, to not be retained, or to be retained only in a specific manner... I don't remember too well, but the article contains some helpful links.
A little off-topic, but...
on
Pirate DNS?
·
· Score: 1
Undernet, the one I'm aware of, was and still is a successful irc network even though they use their own server instead of basic ircd. Also DALnet is another such example, in fact many smaller networks use some version of DALnet's own server. Not only were these successful, but circa 1996 they were both manageably small, while the main network, EFnet, had become the "slums" of IRC due to overpopulation and poor administration. (although DALnet nowadays suffers from both as well)
Yeah... hard to imagine how we will use them all up... but somehow, I figure we will anyway. Somehow, through some sort of inefficiency or multi-ip listening scheme or somesuch... I just don't think we're a society that can have "enough" of any limited resource.:P
> remotely useable because they need to handle the existing searches > of a growing host list and they need to handle new search > bots consuming their bandwidth with "deep scans".
First off, I apologize in advance for posting about GNUtella when I really don't know much at all about it... but I wonder if individuals might at their option run a program or script to index the content and make it available as a single file, which the search bot would specifically retrieve... it seems like that might reduce some overhead. (Esp. if the index file is compressed after being generated)
Glad to find there are still some Gopher servers being maintained. I kinda lost a lot of hope when gopher.tc.umn.edu went down...
Gopher does tend to be a lot more organized, and thus better for the real sharing of information rather than merely presenting things in an aesthetically pleasing manner.
I was going to put a Gopher server up on kuri.weyland-yutani.net and try to maintain links to what Gopher servers were left, but lost my internet connection from a University when I changed schools.:(
I agree that support at the kernel level is best for most hardware. In fact, I still don't understand what hardware vendors feel they lose by giving away the information as to how the device is accessed. I remember using this type of info back on my Apple ][ + to use stuff from BASIC, so it's not exactly a new idea. Given that we now have a push towards standards, it seems vendors would have everything to gain by letting people see this. Given enough demand for the product, Linux drivers would likely be written independently, saving the vendors the trouble of supporting another OS.
Slashdot Mirror
Now what stereotype do I put them under? </joke>
New, improved, smarter lamers?
That being said, it's not much of a stretch to suggest a login tool (for X; those who prefer text environments probably won't mind simply typing guest to start) which can be configured to automatically log in as guest, or to provide a button to do so directly at the login screen. The "guest" account would be disallowed for telnet, ftp, etc. and any security risks would probably be considerably fewer than those present in Windows 9x environments. Actually, I'm a little surprised a program like this hasn't been made yet. (Maybe it has... anyone know of an implementation of something like this?)
In my opinion, user-friendliness issues for Linux generally stem from what hasn't been done yet, not what can't be.
I'm not certain it's really fair to say that there is no legal protection from spam simply because one can't legally say who is allowed to read public documents. Telemarketing and related issues (e.g. UCE) can be considered a form of harrassment. One is free to speak, but that does not mean one can force people to listen to oneself. There are already legal restrictions on telemarketing; there's a good site for info on this, but I've forgotten the address...
Well, all I can say is... You don't have to give them any info. If you tell them you don't want to give them that information, they can (and as a matter of customer service do) continue the transaction without it. I used to work there back at the end of 1997 so this came up a few times.
I believe the currently applicable law you want is the DMCA, which has been discussed on Slashdot (check Microsoft's request to Slashdot regarding the Kerberos-related posts). I haven't finished reading it but I think it required certain specific things either to be retained, to not be retained, or to be retained only in a specific manner... I don't remember too well, but the article contains some helpful links.
Undernet, the one I'm aware of, was and still is a successful irc network even though they use their own server instead of basic ircd. Also DALnet is another such example, in fact many smaller networks use some version of DALnet's own server. Not only were these successful, but circa 1996 they were both manageably small, while the main network, EFnet, had become the "slums" of IRC due to overpopulation and poor administration. (although DALnet nowadays suffers from both as well)
Oh, well, as long as I can have my 256 IPs. :P
> of a growing host list and they need to handle new search
> bots consuming their bandwidth with "deep scans".
First off, I apologize in advance for posting about GNUtella when I really don't know much at all about it... but I wonder if individuals might at their option run a program or script to index the content and make it available as a single file, which the search bot would specifically retrieve... it seems like that might reduce some overhead. (Esp. if the index file is compressed after being generated)
Gopher does tend to be a lot more organized, and thus better for the real sharing of information rather than merely presenting things in an aesthetically pleasing manner.
I was going to put a Gopher server up on kuri.weyland-yutani.net and try to maintain links to what Gopher servers were left, but lost my internet connection from a University when I changed schools. :(
I agree that support at the kernel level is best for most hardware. In fact, I still don't understand what hardware vendors feel they lose by giving away the information as to how the device is accessed. I remember using this type of info back on my Apple ][ + to use stuff from BASIC, so it's not exactly a new idea. Given that we now have a push towards standards, it seems vendors would have everything to gain by letting people see this. Given enough demand for the product, Linux drivers would likely be written independently, saving the vendors the trouble of supporting another OS.