Ummm... so if I'm understanding this right, Apple used TM and C to sue Daewoo and eMachines. And timothy, who posted this, goes off on a rant about how patents can ruin things. I think it's pretty clear that if you design a revolutionary computer, then wholesale copying of that design hurts the innovator, and provides nothing to the customers; look at how many all-in-ones have made it past Apple (all the ones that weren't blattently iMac clones), and then ask why a clone is needed.
Re:Apple, please fix widgets in Classic environmen
on
MacOS X DP3
·
· Score: 1
Not quite the same userbase.
You're right. You're also asking the developers to learn a development system which users no longer interested in buying software aren't interested in. Then again, you're also combining two markets -- I think I can safely claim that damn near every NeXTSTEP/OPENSTEP[for Mach, not necessarily OPENSTEP for NT or Solaris or HP-UX] who can afford to, will switch to MacOS X. So you're losing the people who don't buy software, and gaining a market of people who originally paid $10,000 minimum for a computer (as well as hobbyists such as myself who lusted after $10k machines for years after they came out).
So yeah, there are some differences in the user base -- there's some weeding out and some reseeding -- but it's still a vastly different situation than what Win developers asked to move to OS/2 experienced.
Re:Apple, please fix widgets in Classic environmen
on
MacOS X DP3
·
· Score: 1
Well, Win3.x windows look like Win95 windows when running under Win95 and millions of people still upgraded their apps. WinOS/2 apps looked nothing like OS/2 apps and nobody upgraded to the OS/2 apps.
What OS/2 apps? The difference is, with OS/2, you have to ask the developer to begin development on a new platform with a different market share. With MacOS X, you have to ask the developer to begin development on a new platform with the exact same customer base as before. So yeah, I think the developers will be more responsive.
Who knows? I offered to donate a low-end dual processor system for testing (obviously not an actual development machine, I am a college student), and went ignored. I am under the impression that they want kernel threads working (man kthreads in 2.6 and you'll see what I mean) before they begin work on SMP.
I have three SMP machines at work, though, that I'd love to move to OpenBSD on. Given the performance boosts I've seen on my home machine (which has half the processing power of a single CPU at work, and a quarter the RAM), I'd very much like to see the difference in performance at work:)
NS has this kind of menus because this structure offers better usage then the orginal menu bar idea *if* your screen is large enough. These vertically stacked menus do not waste space anymore than a standard menu bar do.
Nope. It renders the screen space available to users into a shape that doesn't fit the shape of the windows. Since the menu height is variable, there's no set distance down I can put a window so that it doesn't get run over. And don't forget getting the right menu selection takes more time than in MacOS.
And further, the decision to do the menu like that was most certainly not a willing design decision based on the assumption that the way they were doing it was better.
Being able to move then around makes them *very* convenient.
True, but irrelevant to the poor design of the rectangular menu bar.
Just for grins I've compiled a quick look at other operating systems that use something like this:
OS/2 -- LaunchPad / Toolbar. Its shape is definitely annoying. ObjectDesktop unfortunately copies this in some of its tools, but not all (thank God!). At least with version 4, you can elect to use the WarpCenter rather than (or in addition to) the Toolbar; the WarpCenter takes up the entire top or bottom 15 pixels or so, and provides access to a list of running processes (closer to the Macintosh Application menu than the Windows Taskbar, except it lists each process or window seperately), all objects on the desktop, and user-defined trays that hold any objects you damn well please.
BeOS -- is a curious cross between MacOS and Windows. Each window will have its own menu bar, but windows are listed by application in the MacOS-Application-menu-like system menu. This menu increases in size with the number of applications running, but it can be made to run across the top or bottom of the screen and get out of the way.
NeXTSTEP has no root menu. What is the root menu in WindowMaker is the Application menu in NeXTSTEP, in a standard installation it is in the upper right hand corner (it can be moved around, and made to pop up wherever the RMB is clicked, like an application-specific root menu).
According to folklore, NeXTSTEP has that menu system because they were afraid of simply wholesale copying the application menu in MacOS (the one across the top). I can't think of any other reason they'd waste the space with an "always on top" object that can grow both vertically (more options) and horizontally (long words).
And yes, NeXTSTEP originally ran on 17" monitors minimum. The Color NeXTs could come with 17" and 21" monitors, and a NeXTDimension system (NeXTCube with NeXTDimension board) could come with a 17" monochrome and a 21" color monitor, dual-head.
The 64x64 icons seemed quite reasonable on screens of that size, and even the annoying application menu isn't too annoying there. NeXTSTEP for intel suffered in that many people (myself included) stuck it on a system with a 15" monitor, which it was never designed for.
Since the smallest display Apple currently offers is a 15" flat panel (and the smallest CRT is 17"), I suspect that current complaints about the size of Aqua icons (variable, apparently, from 32x32 to 128x128) will be rendered "not too relevant."
Yeah, but the ssh 1.5 protocol (the one implemented by all ssh 1.x clients) has, IIRC, some inherent flaws that allow it to be cracked; further, ssh 1.2.28 doesn't demonstrate the same attention to detail that OpenSSH has demonstrated, like (for instance) the recent RSAREF bug that OpenSSH prevented from being tickled.
OpenSSH, while vulnerable when implementing the 1.5 ssh protocol, also supports an incompatible "1.6" ssh protocol that is immune to the known vulnerabilities of the 1.5 protocol; between OpenSSH clients, this protocol can be spoken instead, improving security.
I'm not great at security, but I try to pay attention to those who are:)
As howardjp pointed out, OpenBSD is not affected (I'm repeating it since I get a +1 Bonus and this will be seen by people, since moderators rarely come here -- if a moderator sees this, please moderate his post, rather than mine, up).
This is exactly the sort of thing that a code audit for security can flush out and fix, so it should come as no surprise that OpenBSD fixed it.
SCSI still supports multi-tasking OS's better. When I'm compiling and using netscape, that means something (although only because of netscape sigh).
Also, if you stick with SCSI drives from good companies (apparently just IBM now:), you'll get reliability improvements over their IDE drives, which are some damned reliable IDE drives.
So, for systems that multitask with disk activity (lots of OSs now), and with drives that typify SCSI, rather than random "I can sell SCSI drives now" vendors, SCSI is still better. Expensive, but worth it.
(Deleted rant on SCSI vs. IDE installation since it was too angry:)
(this is the reason that Jack straw will be getting my encryption keys sometime a few millenia after hell freezes over.
Heh. I'm speaking, however, about a lack of privacy with loved ones, and no one else. If people think I have something to hide because I don't feel they're entitled to something, so be it. If I think they are entitled to something, then I think they are entitled to see me as I am, with out me even tinting the truth.
But for everyone else, stay the Hell out of my mail:)
I frankly believe that the number of things a person doesn't know about you is proportional to how close the two of you are (think secrets=k*ceiling(distance)). But I also think that such trust is predicated on understanding, not a search for proof.
Quite simply, unless you see a person at many levels, you don't know them very well. By knowing my parents' history before I was born, what they did and where they went, etc., I know a lot more about who they are. If you shield a person from certain levels, or certain things, you are setting a boundary for intimacy. Not that that's a bad thing, it's simply something to be recognized -- there is some level of intimacy that you don't want.
Me, I have an open-book policy: my girlfriend has access to all my data when she's around (long distance relationship, and since she doesn't have access to ssh I won't give her my root password), and I answer every question she or my family asks truthfully. My family also has total access to my room and everything in it when I'm not around, but not when I'm there since I need moment-to-moment privacy.
So there's my take on it -- without specific details the (ex-?)girlfriend looks like someone with whom Pete couldn't have that level of intimacy, based on her reaction; and Pete didn't want it anyways.
Just a quick note, if you got your SSH from Replay.com (which provides Red Hat packages), and got the US version, you're using RSAREF. If you got the Debian package, you're not. I'm not sure if Replay.com has updated their RPMs or not, though.
For me, that means the machines at work were vulnerable, and the machines at home are not:)
The OpenSSH implementation checks all outputs it gives, as well as all inputs it gets, so it won't cause the bug.
This is according to Theo himself, who seems far too precise (OK, anal) to let pride get in the way. Me, on the other hand, correcting myself for the second time, am obviously fine spewing crap out. sigh
this bug compromises the security of an OpenBSD system in the US only
Excuse me, I didn't quite get this one right: please insert "if installed before the patched RSAREF was available and not updated since" before "only".
Actually, as I recall the discussion on the OpenBSD mailing list when this came up a week or two ago, this bug compromises the security of an OpenBSD system in the US only (due to our wonderful patent laws). Further, the latest version of OpenBSD (2.6) is not I repeat not affected by this, since it uses OpenSSH -- an implementation of SSH that has undergone their scrutiny.
As for that "urban legend," you're smoking crack. OpenBSD is very up-front that everything in the base installation has undergone extensive testing; ssh, because it is not free software, is not in the base install. Idiot.
I think we should get lawsuit-happy here -- if only because it's our only recourse. That is, everyone in Europe or Oregon who had this happen to them, who reads/., should see about suing Real.
so that the places that currently have [AS]DSL will get VDSL, and the places that don't currently have anything get [AS]DSL... that would mean yours truly gets better bandwidth!
For Microsoft. The way it seems works is, if you're a standard NT user then you have to be licensed; the special consideration is that they're making the same stuff you had available before from another NT box available through the web, or at least they're licensing it like that will be the result.
If you're using NT in the environment Microsoft wants you to use it, this is chickenfeed and pretty typical -- like requiring licenses for people that either telnet or ssh in, rather than not limiting ssh logins. Of course, if you're using NT in the environment Microsoft wants you to use it, you might need that extra money for more support staff and backup machines:)
OK, I've had experience with FreeBSD 2.2.8, 3.1, and OpenBSD 2.3.
So here's my personal opinion: for intel computers that have a single processor, use OpenBSD. For intel processors with more than one CPU, use FreeBSD. For other architectures, you may get to choose between OpenBSD and NetBSD, and I'd say go with OpenBSD first. Why?
First, FreeBSD supports multiple processors, and the other two don't. I'm planning on donating an old SMP pentium to encourage such development, but right now it's not there.
Otherwise, I recommend OpenBSD because its security enhancements are significant. It's FTP server is certainly more secure than wuftpd or Proftpd, and I'm pretty sure it's more secure than the NetBSD and FreeBSD FTP servers. More secure in a 'no longer able to get root through FTP' kind of way, which I would consider significant. Similarly with most everything else; things that today are called theoretical security problems are tomorrow's great exploits in the CERT advisories.
As for NetBSD, there really are a lot of computers you might scoff at, that can run a perfectly acceptable set of small services under NetBSD. And hey, if you can't do "real computing" on that Mac IIci in your closet, you can still use it as a damned fine terminal in whatever room in your house still doesn't have an internet connection. Don't underestimate how many computers you can suddenly find yourself being able to use again, with NetBSD. As far as I know, both FreeBSD's and OpenBSD's linux emulation is great, to the point that huge commercial products (most noticable Netscape and Star Office) run fine; OpenBSD (and possibly NetBSD?) also has a wide collection of OS emulation possibilities, depending on the architecture it's running on (i.e., expect emulation for most UNIX binaries for the given platform, if you tell the kernel so). Similarly, KDE and suchlike UI enhancements run fine, which means that the old saw about the BSD's not being for the desktop is hogwash; administration isn't more difficult, it's just different.
So, aside from my preference for OpenBSD when it's a possibility (and OpenBSD comes from the NetBSD tree, so you can run OpenBSD on that Mac IIci too), go with whatever works.
Sounds like you've already bought the hardware -- sorry about that. You probably should have considered a Mac (PowerBook or iBook if grandpa doesn't mind the color), since there is considerably less to worry about there. If not MacOS, consider Linux CLI, or WindowMaker, or Afterstep (too bad NeXT Computers never made a laptop...).
For those people suggesting he go with Windows: no. There's a hell of a lot more to learn in using Windows graphically than in using X graphically, since Windows has a richer and more in-depth metaphor to work from. If the guy is 89 years old, he doesn't have a lot of time to learn the desktop paradigm, and once Linux and X are set up it's pretty damned simple. Click or double-click (one click for afterstep, I think?) on the picture over on the right side of the screen, your program pops up. Click on the program's top edge to move it around, click on another program to see the whole program. In the apps he's likely to use, Using the File menu to exit would work and and the little buttons on the top of the program can safely be ignored. In fact, most everything can be safely ignored, and papers can be typed and the web can be surfed.
Slashdot Mirror
Ummm... so if I'm understanding this right, Apple used TM and C to sue Daewoo and eMachines. And timothy, who posted this, goes off on a rant about how patents can ruin things. I think it's pretty clear that if you design a revolutionary computer, then wholesale copying of that design hurts the innovator, and provides nothing to the customers; look at how many all-in-ones have made it past Apple (all the ones that weren't blattently iMac clones), and then ask why a clone is needed.
So yeah, there are some differences in the user base -- there's some weeding out and some reseeding -- but it's still a vastly different situation than what Win developers asked to move to OS/2 experienced.
I have three SMP machines at work, though, that I'd love to move to OpenBSD on. Given the performance boosts I've seen on my home machine (which has half the processing power of a single CPU at work, and a quarter the RAM), I'd very much like to see the difference in performance at work
And further, the decision to do the menu like that was most certainly not a willing design decision based on the assumption that the way they were doing it was better. True, but irrelevant to the poor design of the rectangular menu bar.
Just for grins I've compiled a quick look at other operating systems that use something like this:
NeXTSTEP has no root menu. What is the root menu in WindowMaker is the Application menu in NeXTSTEP, in a standard installation it is in the upper right hand corner (it can be moved around, and made to pop up wherever the RMB is clicked, like an application-specific root menu).
According to folklore, NeXTSTEP has that menu system because they were afraid of simply wholesale copying the application menu in MacOS (the one across the top). I can't think of any other reason they'd waste the space with an "always on top" object that can grow both vertically (more options) and horizontally (long words).
And yes, NeXTSTEP originally ran on 17" monitors minimum. The Color NeXTs could come with 17" and 21" monitors, and a NeXTDimension system (NeXTCube with NeXTDimension board) could come with a 17" monochrome and a 21" color monitor, dual-head.
The 64x64 icons seemed quite reasonable on screens of that size, and even the annoying application menu isn't too annoying there. NeXTSTEP for intel suffered in that many people (myself included) stuck it on a system with a 15" monitor, which it was never designed for.
Since the smallest display Apple currently offers is a 15" flat panel (and the smallest CRT is 17"), I suspect that current complaints about the size of Aqua icons (variable, apparently, from 32x32 to 128x128) will be rendered "not too relevant."
Yeah, but the ssh 1.5 protocol (the one implemented by all ssh 1.x clients) has, IIRC, some inherent flaws that allow it to be cracked; further, ssh 1.2.28 doesn't demonstrate the same attention to detail that OpenSSH has demonstrated, like (for instance) the recent RSAREF bug that OpenSSH prevented from being tickled.
:)
OpenSSH, while vulnerable when implementing the 1.5 ssh protocol, also supports an incompatible "1.6" ssh protocol that is immune to the known vulnerabilities of the 1.5 protocol; between OpenSSH clients, this protocol can be spoken instead, improving security.
I'm not great at security, but I try to pay attention to those who are
As howardjp pointed out, OpenBSD is not affected (I'm repeating it since I get a +1 Bonus and this will be seen by people, since moderators rarely come here -- if a moderator sees this, please moderate his post, rather than mine, up).
This is exactly the sort of thing that a code audit for security can flush out and fix, so it should come as no surprise that OpenBSD fixed it.
SCSI still supports multi-tasking OS's better. When I'm compiling and using netscape, that means something (although only because of netscape sigh).
:), you'll get reliability improvements over their IDE drives, which are some damned reliable IDE drives.
:)
Also, if you stick with SCSI drives from good companies (apparently just IBM now
So, for systems that multitask with disk activity (lots of OSs now), and with drives that typify SCSI, rather than random "I can sell SCSI drives now" vendors, SCSI is still better. Expensive, but worth it.
(Deleted rant on SCSI vs. IDE installation since it was too angry
Bah
But for everyone else, stay the Hell out of my mail
I frankly believe that the number of things a person doesn't know about you is proportional to how close the two of you are (think secrets=k*ceiling(distance)). But I also think that such trust is predicated on understanding, not a search for proof.
Quite simply, unless you see a person at many levels, you don't know them very well. By knowing my parents' history before I was born, what they did and where they went, etc., I know a lot more about who they are. If you shield a person from certain levels, or certain things, you are setting a boundary for intimacy. Not that that's a bad thing, it's simply something to be recognized -- there is some level of intimacy that you don't want.
Me, I have an open-book policy: my girlfriend has access to all my data when she's around (long distance relationship, and since she doesn't have access to ssh I won't give her my root password), and I answer every question she or my family asks truthfully. My family also has total access to my room and everything in it when I'm not around, but not when I'm there since I need moment-to-moment privacy.
So there's my take on it -- without specific details the (ex-?)girlfriend looks like someone with whom Pete couldn't have that level of intimacy, based on her reaction; and Pete didn't want it anyways.
Just a quick note, if you got your SSH from Replay.com (which provides Red Hat packages), and got the US version, you're using RSAREF. If you got the Debian package, you're not. I'm not sure if Replay.com has updated their RPMs or not, though.
:)
For me, that means the machines at work were vulnerable, and the machines at home are not
If it were Theo he would have gotten it right.
Actually it isn't :)
The OpenSSH implementation checks all outputs it gives, as well as all inputs it gets, so it won't cause the bug.
This is according to Theo himself, who seems far too precise (OK, anal) to let pride get in the way. Me, on the other hand, correcting myself for the second time, am obviously fine spewing crap out. sigh
Actually, as I recall the discussion on the OpenBSD mailing list when this came up a week or two ago, this bug compromises the security of an OpenBSD system in the US only (due to our wonderful patent laws). Further, the latest version of OpenBSD (2.6) is not I repeat not affected by this, since it uses OpenSSH -- an implementation of SSH that has undergone their scrutiny.
As for that "urban legend," you're smoking crack. OpenBSD is very up-front that everything in the base installation has undergone extensive testing; ssh, because it is not free software, is not in the base install. Idiot.
Cody, is that you? That has to be you. I can't imagine anyone else going by a variation of Judge talking about his 35" TV and 19" monitor...
:)
If I'm wrong, I apologize
I think we should get lawsuit-happy here -- if only because it's our only recourse. That is, everyone in Europe or Oregon who had this happen to them, who reads /., should see about suing Real.
Bah.
so that the places that currently have [AS]DSL will get VDSL, and the places that don't currently have anything get [AS]DSL... that would mean yours truly gets better bandwidth!
:)
Which, after all, is all that matters, right?
For Microsoft. The way it seems works is, if you're a standard NT user then you have to be licensed; the special consideration is that they're making the same stuff you had available before from another NT box available through the web, or at least they're licensing it like that will be the result.
:)
If you're using NT in the environment Microsoft wants you to use it, this is chickenfeed and pretty typical -- like requiring licenses for people that either telnet or ssh in, rather than not limiting ssh logins. Of course, if you're using NT in the environment Microsoft wants you to use it, you might need that extra money for more support staff and backup machines
OK, I've had experience with FreeBSD 2.2.8, 3.1, and OpenBSD 2.3.
So here's my personal opinion: for intel computers that have a single processor, use OpenBSD. For intel processors with more than one CPU, use FreeBSD. For other architectures, you may get to choose between OpenBSD and NetBSD, and I'd say go with OpenBSD first. Why?
First, FreeBSD supports multiple processors, and the other two don't. I'm planning on donating an old SMP pentium to encourage such development, but right now it's not there.
Otherwise, I recommend OpenBSD because its security enhancements are significant. It's FTP server is certainly more secure than wuftpd or Proftpd, and I'm pretty sure it's more secure than the NetBSD and FreeBSD FTP servers. More secure in a 'no longer able to get root through FTP' kind of way, which I would consider significant. Similarly with most everything else; things that today are called theoretical security problems are tomorrow's great exploits in the CERT advisories.
As for NetBSD, there really are a lot of computers you might scoff at, that can run a perfectly acceptable set of small services under NetBSD. And hey, if you can't do "real computing" on that Mac IIci in your closet, you can still use it as a damned fine terminal in whatever room in your house still doesn't have an internet connection. Don't underestimate how many computers you can suddenly find yourself being able to use again, with NetBSD. As far as I know, both FreeBSD's and OpenBSD's linux emulation is great, to the point that huge commercial products (most noticable Netscape and Star Office) run fine; OpenBSD (and possibly NetBSD?) also has a wide collection of OS emulation possibilities, depending on the architecture it's running on (i.e., expect emulation for most UNIX binaries for the given platform, if you tell the kernel so). Similarly, KDE and suchlike UI enhancements run fine, which means that the old saw about the BSD's not being for the desktop is hogwash; administration isn't more difficult, it's just different.
So, aside from my preference for OpenBSD when it's a possibility (and OpenBSD comes from the NetBSD tree, so you can run OpenBSD on that Mac IIci too), go with whatever works.
Sounds like you've already bought the hardware -- sorry about that. You probably should have considered a Mac (PowerBook or iBook if grandpa doesn't mind the color), since there is considerably less to worry about there. If not MacOS, consider Linux CLI, or WindowMaker, or Afterstep (too bad NeXT Computers never made a laptop...).
For those people suggesting he go with Windows: no. There's a hell of a lot more to learn in using Windows graphically than in using X graphically, since Windows has a richer and more in-depth metaphor to work from. If the guy is 89 years old, he doesn't have a lot of time to learn the desktop paradigm, and once Linux and X are set up it's pretty damned simple. Click or double-click (one click for afterstep, I think?) on the picture over on the right side of the screen, your program pops up. Click on the program's top edge to move it around, click on another program to see the whole program. In the apps he's likely to use, Using the File menu to exit would work and and the little buttons on the top of the program can safely be ignored. In fact, most everything can be safely ignored, and papers can be typed and the web can be surfed.
Except, of course, that real X terminals are cheaper... just not as cool looking :)