Its a shame that a version of the braindead DOS command line lives on in modern versions of Windows and hasnt been replaced with something closer to what Unix has.
What's surprising is that DOS *hasn't* been replaced by something better and more similar to the shells available under Unix. One of the first things people talk about as being reasons to use UNIX over Windows is the power and flexibility of the shell.
At the very least I would have expected something more sh(1)-like, even if it did choose to include a lot of older MS-DOS commands. At the most I would have expected something that was *compatible* with sh(1) with a lot of the extensions from bash or zsh that people have come to expect, along with the kinds of things that would make it useful in a Windows GUI environment, like some *very* basic GUI dialog features that could prompt for yes/no or single line input without a invoking a cmd shell, but no complex windowing behavior or event-driven programming.
MS has responded with the "improved" features of the NT command shell and Windows scripting (which I presume is a VB script derivative), without realizing that DOS batch file compatibility isn't terribly helpful and complex VBScript and GUI interaction won't get used.
People, especially admins, want a fair amount of power (loops, variables, substitions, output redirection, etc) and no complex GUI interaction or dependencies. But they want security and stability, too, and MS hasn't always made it a priority to deliver those features either...
I'm sure that most people will disagree with me, but I think not having to upgrade your computer to get an equivilent experience as a person with a new PC is a fairly recent phenomena.
The key thing is "equivilent experience" -- sure, you can browse the web and send email on a 386 with 16MB of RAM running Linux, Lynx and Pine, but its not the same experience that a person running a newer system with a GUI, new browser, plugins, etc. I'd argue that an absolute bottom of the barrel equivilent experience would to have to be 98SE/ME on a PII450 with 256MB of RAM. Anything below that just isn't the same as P4 running XP.
Sure, there are some Linux trolls out there happy to deal with sluggish old P1s and P2s, but they're not getting the same experience.
I don't really notice a difference with my "old" computer (2.5 yr old dual PIII, WinXP) and brand-new P4s with XP. But had this been 4 years ago and I was trying to run Win2K Pro on a P1 166, it would have been glaringly obvious (yes, I have done this).
I'd attribute most of the comparability between 2-3 year old systems and new systems to the lack of overwhelming mobo throughput increases but mostly to the relative OS stability over the last three years -- the economic slowdown has definitely prompted MS to slow its OS upgrade cycle a little.
Wouldn't it make more economic and performance sense to cut the number of PCs by a third and take the $50k and invest in a more high-performance and space-conservative disk subsystem?
The FTC, states' Attorneys General and other crimefighting organizations need to start going after the fraud that's behind almost every single SPAM message I've ever seen. The spammers (the people sending the email) are almost always hard to get to, but in order for the whole thing to be worth it there must be some way to get to the sellers, otherwise they couldn't collect money from the rubes that reply.
Why can't we get law enforcement to start nailing the scam artists responsible for the spam being generated in the first place? I mean, putting guys in *jail*, big civil fines, and so on.
We can bitch all we want about the clowns sending email, but if the fraudsters were starting to get locked up on a frequent, regular basis it would dry up the market for spammers and they'd move on to something else.
AND if we bitch too long about spam, we're liable to end up with some icky government mandated "system" about email -- how would you like to have to get a license from the government to run an email service? It's to prevent spam, you know...
Some businesses appear to be able to kick people out or not let them in for any reason.
I'd suspect that nightclubs would claim that they are private clubs, not public places. Casinos probably have laws specific to them on the books that allow them to kick out anyone "cheating."
I really don't know how stores prevent people from coming back who have shoplifted. It may be the case that they just get shadowed about 6" away by a couple of loss-prevention people the entire time they're there, making the experience unpleasant for the shoplifter.
There's clearly something that prevents you from arbitrarily kicking people out of a place that is truly a "public business", otherwise lots of places would have an unwritten racial policies.
Of course, there's the downside that I don't get to use the meeting-scheduling interfaces
I've long wondered why the IMAP protocol hasn't been extended to allow for meeting-scheduling and calendaring functionality to be built into any IMAP client.
Most of it would be to support real-time busy search, but the rest of it could just be storing calendar items as ordinary messages in a standardized format that could be parsed by the client and displayed any way the client app deems approrpriate.
This would really cool, as it would allow any IMAP client that supported "IMAP calendaring extensions" to be able to get pretty much full calendaring functionality out of any IMAP server that supported them as well; pretty much any server could then have complete calendaring abilities.
It would also break the "but we need calendaring" stranglehold that Exchange has on messaging.
The big missing element would be multi-server busy search; you'd have to have a way for IMAP servers to communicate with each other. You'd also need a decent LDAP directory, but that would be icing and not totally necessary for functionality.
If they have the IMAP stuff turned on for Exchange 2k, you can run Pine as an IMAP client against Exchange. I do this when I'm suspicous about a mail message and don't trust Outlook.
It's kind of amusing to do it, and more functional than using lynx with Outlook Web Access..
Well, my "storage technique" is deliberately designed to falsify ownership of property. To own property you have to do a bunch of government paperwork to update the county land rolls, tax rolls, etc. Deliberately lying on those forms has GOT to be illegal.
The biggest downside would be the suspicion that would come from the locals. "Who's this guy and why did he buy a parcel of land that's just a bunch of woods"?
2) Buy plot of land in extremely rural area close to Canadian border. Use false identity, pay cash.
3) Build small, subterranean concrete bunker (10' x 10'). Install water-tight safe in bunker. Camouflage bunker, make it tamper-evident.
4) Visit with data periodically.
You now have a safe place to store things. Safe from fire, flood, and most importantly from the government. Since you bought the land with false identification, they can't shake you down for what you have stored there, unless they know about it. It's close to the border, so you should be able to get the contents fairly easily from the other side of the border -- or get the data as you go OVER the border.
OK, so its not convenient and illegal, but hasn't true safety and privacy always been that way?
It surprises me too. I'm surprised that given the number of really smart people at Microsoft and the obvious things that people like about UNIX that they haven't found a way to at least clone if not openly co-opt the functionality of much of UNIX:
Provide a new shell that happens to run scripts written for/bin/sh perfectly. Better process management. More transparent process control and dependency management. Improved security.
I imagine meetings like in "Office Space" -- The Two Bobs and Lumberg listening to the bright engineers suggestions, the engineers leave, and then the managers come out and discuss doing the exact opposite.
Unless you really do believe that there is a deliberate conspiracy on the part of Microsoft to make a shoddier product because they can and to just market the shit out of it, it's hard to understand why MS does some of the things they do..
Heh, this is slashdot, where, to mangle a quote from Lenin "rhetoric is a fact of its own".
I worked in a video store that had a problem with what I'd call vandalism or nuisance behaviors -- two specific kids would come in, tear around in circles, knocking over movie boxes, throwing shit until we chased them out.
After chasing them out several times in one day we called the police who said "they're not trespassing, it's a place of public business, there's little we can do".
The cops did show though, and they put them in their car and drive them home (bad neighborhood, bad scene, you know the tale), and they only came back once, but we were REAL firm with them -- we physically ejected them from the store via the rear exit, an alleyway which meant a 2 block walk to get back to the front entrance.
And when I say physically ejected, I mean physical. The larger (maybe 12?) kid I put in a hammerlock and a chokehold and threw out the back door, kicking him in the ass as he stumbled out. The smaller kid (10?) was fireman carried out the back of the store. Both were warned out back of a severe beating if they ever came back again.
I expected to get shot or stabbed in the next day or two, but nothing ever happened. I suspect that in most non-corporate stores, this is the treatment you get if you're too much of an asshole.
I've gotten spam like that to accounts that have zero usage or knowledge by others.
I've assumed that:
1) Spammers do random generation sends
2) Spammers harvest the left hand side of email addresses and the hit big ISPs with the dictionary of usernames @bigISP.com. This would make sense, since lots of people are free with their username on web forums or usenet, assuming that guarding the RHS of their address is enough.
3) ISPs and service providers that claim they will never spam are lying or at least internally rationalizing the stretched-to-breaking privacy policies so they can sell email addresses.
Heh, I'm an admin too, and I feel your anger -- especially when management insists on shedding jobs to "save money" and the jobs they shed are all the good people, leaving nothing but the fools and incompetants.
In fact, the fools here are so stupid and lazy, not only can they not see the value of an older system as an Xterm or Winterm, they don't have the brainpower to call Dell to get the new systems fixed while under warranty.
But the labor problem in IT is really a management problem, as management doesn't really understand IT in most industries and simply looks at it as an overhead cost without calculating the savings IT brings to the table.
Hiring smart IT people with broad skills, presenting the business challenges to them and giving them latitude to solve them seems so obvious, but they instead choose to hire the dumbest labor they can and either ignore all the waste and opportunity costs, or use those costs as a justification for dumbing IT down even further.
I wasn't trying to start a war about how much open source costs, I was just making a point that it's a bigger and deeper argument than just license costs.
everyone here already knows that commercial software is more expensive.
More expensive to buy, but as I'm sure Microsoft or other commercial software vendors would argue, is it more expensive over its life cycle?
If you spent $10k on licenses and hired a low-end MS monkey for $35k, you'd be out $45k. If you spent zero on licenses, you'd likely have to pay a more highly skilled admin $50k, costing you even more than the commercial software option even though you didn't pay anything for the software itself.
It's an overly simplistic argument, I know, and there are responses regarding reliability, security, hardware upgrades, and so on, but I think there are reasonable arguments to be made about the cost of commercial software versus the non-obvious costs of free software.
...or more specifically transfers, when do I get my money? I mean I got your email and everything and I'm more than willing to help you get your, I mean our, money out of the country.
However, their store is private property. They can ask you to leave any time they want, for any reason.
Really? So if I own a store that's generally open to the public (glass windows and doors, sign on the door announcing hours, store name, etc) I can walk up and say "Whites only -- leave now" and actually get away with it?
CDs are WAY overpriced. A new release DVD (single disc) is generally about what, $20? A new release CD is around $15. I think a movie takes longer and costs more to make -- but why is the DVD priced so closely to the CD? Only reasons I can think of I can debunk myself.
CDs cost as much to make as movies. False. It would seem really unusual for an artist to spend $20M recording a CD. Many movies cost $40-50M to make.
Movies pay off their costs at theaters. False. Most movies don't even break even, they rely on rental and sales to break even or profit.
A person in such a situation could look at case studies, talk to sysadmins at other companies as well as friends, they could also look through the plethera of books and make an educated decision. This decision could then be presented to management.
In other words, be a sysadmin? If it walks like a duck, talks like a duck, then it must be a duck.
Maybe what I should have read was "I'm a sysadmin, but I don't have any confidence."
Granted, he was working within the industry, but the devastating piracy figures in a recent poll conducted among computer users made it clear that DRM will save the industry a lot of money.
How is piracy even costing them money? Even the assumption that each pirated copy is lost revenue is a fallacy since it assumes that all the pirated copies would have been purchased if the pirate copy wasn't available.
The assumption that pirated copies of anything cause the seller to accrue costs is pure fantasy, unless you're disingenuous enough to include the cost of their antipiracy efforts.
The fact is that copying content doesn't cost the original issuer anything. I'd only grant them about 10% lost sales due to piracy -- many people who pirate just get it because they can, not because they want it.
Don't think that most companies are "hoodwinked" by people who want to ride the "Linux Train". What does that mean, anyway?
I just thought it was an interesting contradiction that the parent poster somewhere above was complaining about how they didn't know very much about being a sysadmin, but they had convinced management to totally switch to another platform.
What did they convince management with? Since they don't know very much, it wasn't facts, it was rhetoric about how great Linux is.
Yes, using fdisk and disklabel to partition new disks is difficult for a newcomer, but actually, across all the BSD's that's pretty much "standard". You did it once on FreeBSD, well, you won't have to relearn much for Net and OpenBSD. Great huh?
God I hate that rationale. "It's broken and braindead across other platforms, too, so you should be able to deal with it there, too".
I agree that such things must become somewhat less contused. But the BSD's use a wholly different systems than the intel partition map
cruft that e.g. Linux uses.
What's specifically wrong with it, besides the fact that its the same way MS does it? I'd wager that its simpler and solves the same problem just as well.
There are actually quite a few benefits in making one big slice and partitioning that up. You can get lots more partitions on a disk that way. Sysinstall is a step in the right direction for making this easier, but it's not perfect.
I have a hard time seeing the need for a zillion partitions. Maybe 6, tops -- there's no performance increases, and its certainly an inflexible way of allocating disk space.
I just wish there was a linux-style fdisk application (simpler, easier) that did everything that the disk manipulation portion of sysinstall did, but simpler.
I've been looking for a reliable way to do this for some time and haven't found one.
They all seem built around the idea of local delivery, which is fine, but I'm interested in doing it on a pure mail hub that doesn't do local delivery.
The closest I've seen is spamass-milter, but it won't build on FreeBSD currently as it requires Autoconf 2.53, which happens to be marked as broken right now.
That might be true, but there are countless companies that have NO sysadmin. You have, instead, an overworked project manager who has been forced to work on IIS w/ some piece of crap Servlet Container for the last umpteen years. Now that he's convinced management to let him run linux and tomcat w/ apache he has very little time to set it up, and not a ton of experience.
He's not a sysadmin, but knows enough to convince management to let him run linux and tomcat and apache.
Is management that gullible that they can be hoodwinked by someone who isn't a sysadmin, but wants to ride on the linux train?
Slashdot Mirror
Its a shame that a version of the braindead DOS command line lives on in modern versions of Windows and hasnt been replaced with something closer to what Unix has.
What's surprising is that DOS *hasn't* been replaced by something better and more similar to the shells available under Unix. One of the first things people talk about as being reasons to use UNIX over Windows is the power and flexibility of the shell.
At the very least I would have expected something more sh(1)-like, even if it did choose to include a lot of older MS-DOS commands. At the most I would have expected something that was *compatible* with sh(1) with a lot of the extensions from bash or zsh that people have come to expect, along with the kinds of things that would make it useful in a Windows GUI environment, like some *very* basic GUI dialog features that could prompt for yes/no or single line input without a invoking a cmd shell, but no complex windowing behavior or event-driven programming.
MS has responded with the "improved" features of the NT command shell and Windows scripting (which I presume is a VB script derivative), without realizing that DOS batch file compatibility isn't terribly helpful and complex VBScript and GUI interaction won't get used.
People, especially admins, want a fair amount of power (loops, variables, substitions, output redirection, etc) and no complex GUI interaction or dependencies. But they want security and stability, too, and MS hasn't always made it a priority to deliver those features either...
I'm sure that most people will disagree with me, but I think not having to upgrade your computer to get an equivilent experience as a person with a new PC is a fairly recent phenomena.
The key thing is "equivilent experience" -- sure, you can browse the web and send email on a 386 with 16MB of RAM running Linux, Lynx and Pine, but its not the same experience that a person running a newer system with a GUI, new browser, plugins, etc. I'd argue that an absolute bottom of the barrel equivilent experience would to have to be 98SE/ME on a PII450 with 256MB of RAM. Anything below that just isn't the same as P4 running XP.
Sure, there are some Linux trolls out there happy to deal with sluggish old P1s and P2s, but they're not getting the same experience.
I don't really notice a difference with my "old" computer (2.5 yr old dual PIII, WinXP) and brand-new P4s with XP. But had this been 4 years ago and I was trying to run Win2K Pro on a P1 166, it would have been glaringly obvious (yes, I have done this).
I'd attribute most of the comparability between 2-3 year old systems and new systems to the lack of overwhelming mobo throughput increases but mostly to the relative OS stability over the last three years -- the economic slowdown has definitely prompted MS to slow its OS upgrade cycle a little.
Wouldn't it make more economic and performance sense to cut the number of PCs by a third and take the $50k and invest in a more high-performance and space-conservative disk subsystem?
Something like this.
Would give you far better disk performance and scalability than trying to add another 200 PCs with IDE disks.
3000 miles of shelf space -- at what type size?
Would the Large Type Edition be [cue Mr. Evil voice]:
One Million Miles!?!
The FTC, states' Attorneys General and other crimefighting organizations need to start going after the fraud that's behind almost every single SPAM message I've ever seen. The spammers (the people sending the email) are almost always hard to get to, but in order for the whole thing to be worth it there must be some way to get to the sellers, otherwise they couldn't collect money from the rubes that reply.
Why can't we get law enforcement to start nailing the scam artists responsible for the spam being generated in the first place? I mean, putting guys in *jail*, big civil fines, and so on.
We can bitch all we want about the clowns sending email, but if the fraudsters were starting to get locked up on a frequent, regular basis it would dry up the market for spammers and they'd move on to something else.
AND if we bitch too long about spam, we're liable to end up with some icky government mandated "system" about email -- how would you like to have to get a license from the government to run an email service? It's to prevent spam, you know...
Some businesses appear to be able to kick people out or not let them in for any reason.
I'd suspect that nightclubs would claim that they are private clubs, not public places. Casinos probably have laws specific to them on the books that allow them to kick out anyone "cheating."
I really don't know how stores prevent people from coming back who have shoplifted. It may be the case that they just get shadowed about 6" away by a couple of loss-prevention people the entire time they're there, making the experience unpleasant for the shoplifter.
There's clearly something that prevents you from arbitrarily kicking people out of a place that is truly a "public business", otherwise lots of places would have an unwritten racial policies.
Of course, there's the downside that I don't get to use the meeting-scheduling interfaces
I've long wondered why the IMAP protocol hasn't been extended to allow for meeting-scheduling and calendaring functionality to be built into any IMAP client.
Most of it would be to support real-time busy search, but the rest of it could just be storing calendar items as ordinary messages in a standardized format that could be parsed by the client and displayed any way the client app deems approrpriate.
This would really cool, as it would allow any IMAP client that supported "IMAP calendaring extensions" to be able to get pretty much full calendaring functionality out of any IMAP server that supported them as well; pretty much any server could then have complete calendaring abilities.
It would also break the "but we need calendaring" stranglehold that Exchange has on messaging.
The big missing element would be multi-server busy search; you'd have to have a way for IMAP servers to communicate with each other. You'd also need a decent LDAP directory, but that would be icing and not totally necessary for functionality.
If they have the IMAP stuff turned on for Exchange 2k, you can run Pine as an IMAP client against Exchange. I do this when I'm suspicous about a mail message and don't trust Outlook.
It's kind of amusing to do it, and more functional than using lynx with Outlook Web Access..
Well, my "storage technique" is deliberately designed to falsify ownership of property. To own property you have to do a bunch of government paperwork to update the county land rolls, tax rolls, etc. Deliberately lying on those forms has GOT to be illegal.
The biggest downside would be the suspicion that would come from the locals. "Who's this guy and why did he buy a parcel of land that's just a bunch of woods"?
1) Create false identity
2) Buy plot of land in extremely rural area close to Canadian border. Use false identity, pay cash.
3) Build small, subterranean concrete bunker (10' x 10'). Install water-tight safe in bunker. Camouflage bunker, make it tamper-evident.
4) Visit with data periodically.
You now have a safe place to store things. Safe from fire, flood, and most importantly from the government. Since you bought the land with false identification, they can't shake you down for what you have stored there, unless they know about it. It's close to the border, so you should be able to get the contents fairly easily from the other side of the border -- or get the data as you go OVER the border.
OK, so its not convenient and illegal, but hasn't true safety and privacy always been that way?
It surprises me too. I'm surprised that given the number of really smart people at Microsoft and the obvious things that people like about UNIX that they haven't found a way to at least clone if not openly co-opt the functionality of much of UNIX:
/bin/sh perfectly. Better process management. More transparent process control and dependency management. Improved security.
Provide a new shell that happens to run scripts written for
I imagine meetings like in "Office Space" -- The Two Bobs and Lumberg listening to the bright engineers suggestions, the engineers leave, and then the managers come out and discuss doing the exact opposite.
Unless you really do believe that there is a deliberate conspiracy on the part of Microsoft to make a shoddier product because they can and to just market the shit out of it, it's hard to understand why MS does some of the things they do..
Heh, this is slashdot, where, to mangle a quote from Lenin "rhetoric is a fact of its own".
I worked in a video store that had a problem with what I'd call vandalism or nuisance behaviors -- two specific kids would come in, tear around in circles, knocking over movie boxes, throwing shit until we chased them out.
After chasing them out several times in one day we called the police who said "they're not trespassing, it's a place of public business, there's little we can do".
The cops did show though, and they put them in their car and drive them home (bad neighborhood, bad scene, you know the tale), and they only came back once, but we were REAL firm with them -- we physically ejected them from the store via the rear exit, an alleyway which meant a 2 block walk to get back to the front entrance.
And when I say physically ejected, I mean physical. The larger (maybe 12?) kid I put in a hammerlock and a chokehold and threw out the back door, kicking him in the ass as he stumbled out. The smaller kid (10?) was fireman carried out the back of the store. Both were warned out back of a severe beating if they ever came back again.
I expected to get shot or stabbed in the next day or two, but nothing ever happened. I suspect that in most non-corporate stores, this is the treatment you get if you're too much of an asshole.
I've gotten spam like that to accounts that have zero usage or knowledge by others.
I've assumed that:
1) Spammers do random generation sends
2) Spammers harvest the left hand side of email addresses and the hit big ISPs with the dictionary of usernames @bigISP.com. This would make sense, since lots of people are free with their username on web forums or usenet, assuming that guarding the RHS of their address is enough.
3) ISPs and service providers that claim they will never spam are lying or at least internally rationalizing the stretched-to-breaking privacy policies so they can sell email addresses.
Heh, I'm an admin too, and I feel your anger -- especially when management insists on shedding jobs to "save money" and the jobs they shed are all the good people, leaving nothing but the fools and incompetants.
In fact, the fools here are so stupid and lazy, not only can they not see the value of an older system as an Xterm or Winterm, they don't have the brainpower to call Dell to get the new systems fixed while under warranty.
But the labor problem in IT is really a management problem, as management doesn't really understand IT in most industries and simply looks at it as an overhead cost without calculating the savings IT brings to the table.
Hiring smart IT people with broad skills, presenting the business challenges to them and giving them latitude to solve them seems so obvious, but they instead choose to hire the dumbest labor they can and either ignore all the waste and opportunity costs, or use those costs as a justification for dumbing IT down even further.
I wasn't trying to start a war about how much open source costs, I was just making a point that it's a bigger and deeper argument than just license costs.
everyone here already knows that commercial software is more expensive.
More expensive to buy, but as I'm sure Microsoft or other commercial software vendors would argue, is it more expensive over its life cycle?
If you spent $10k on licenses and hired a low-end MS monkey for $35k, you'd be out $45k. If you spent zero on licenses, you'd likely have to pay a more highly skilled admin $50k, costing you even more than the commercial software option even though you didn't pay anything for the software itself.
It's an overly simplistic argument, I know, and there are responses regarding reliability, security, hardware upgrades, and so on, but I think there are reasonable arguments to be made about the cost of commercial software versus the non-obvious costs of free software.
...or more specifically transfers, when do I get my money? I mean I got your email and everything and I'm more than willing to help you get your, I mean our, money out of the country.
However, their store is private property. They can ask you to leave any time they want, for any reason.
Really? So if I own a store that's generally open to the public (glass windows and doors, sign on the door announcing hours, store name, etc) I can walk up and say "Whites only -- leave now" and actually get away with it?
CDs are WAY overpriced. A new release DVD (single disc) is generally about what, $20? A new release CD is around $15. I think a movie takes longer and costs more to make -- but why is the DVD priced so closely to the CD? Only reasons I can think of I can debunk myself.
CDs cost as much to make as movies. False. It would seem really unusual for an artist to spend $20M recording a CD. Many movies cost $40-50M to make.
Movies pay off their costs at theaters. False. Most movies don't even break even, they rely on rental and sales to break even or profit.
A person in such a situation could look at case studies, talk to sysadmins at other companies as well as friends, they could also look through the plethera of books and make an educated decision. This decision could then be presented to management.
In other words, be a sysadmin? If it walks like a duck, talks like a duck, then it must be a duck.
Maybe what I should have read was "I'm a sysadmin, but I don't have any confidence."
Granted, he was working within the industry, but the devastating piracy figures in a recent poll conducted among computer users made it clear that DRM will save the industry a lot of money.
How is piracy even costing them money? Even the assumption that each pirated copy is lost revenue is a fallacy since it assumes that all the pirated copies would have been purchased if the pirate copy wasn't available.
The assumption that pirated copies of anything cause the seller to accrue costs is pure fantasy, unless you're disingenuous enough to include the cost of their antipiracy efforts.
The fact is that copying content doesn't cost the original issuer anything. I'd only grant them about 10% lost sales due to piracy -- many people who pirate just get it because they can, not because they want it.
Don't think that most companies are "hoodwinked" by people who want to ride the "Linux Train". What does that mean, anyway?
I just thought it was an interesting contradiction that the parent poster somewhere above was complaining about how they didn't know very much about being a sysadmin, but they had convinced management to totally switch to another platform.
What did they convince management with? Since they don't know very much, it wasn't facts, it was rhetoric about how great Linux is.
Yes, using fdisk and disklabel to partition new disks is difficult for a newcomer, but actually, across all the BSD's that's pretty much "standard". You did it once on FreeBSD, well, you won't have to relearn much for Net and OpenBSD. Great huh?
God I hate that rationale. "It's broken and braindead across other platforms, too, so you should be able to deal with it there, too".
I agree that such things must become somewhat less contused. But the BSD's use a wholly different systems than the intel partition map cruft that e.g. Linux uses.
What's specifically wrong with it, besides the fact that its the same way MS does it? I'd wager that its simpler and solves the same problem just as well.
There are actually quite a few benefits in making one big slice and partitioning that up. You can get lots more partitions on a disk that way. Sysinstall is a step in the right direction for making this easier, but it's not perfect.
I have a hard time seeing the need for a zillion partitions. Maybe 6, tops -- there's no performance increases, and its certainly an inflexible way of allocating disk space.
I just wish there was a linux-style fdisk application (simpler, easier) that did everything that the disk manipulation portion of sysinstall did, but simpler.
I've been looking for a reliable way to do this for some time and haven't found one.
They all seem built around the idea of local delivery, which is fine, but I'm interested in doing it on a pure mail hub that doesn't do local delivery.
The closest I've seen is spamass-milter, but it won't build on FreeBSD currently as it requires Autoconf 2.53, which happens to be marked as broken right now.
That might be true, but there are countless companies that have NO sysadmin. You have, instead, an overworked project manager who has been forced to work on IIS w/ some piece of crap Servlet Container for the last umpteen years. Now that he's convinced management to let him run linux and tomcat w/ apache he has very little time to set it up, and not a ton of experience.
He's not a sysadmin, but knows enough to convince management to let him run linux and tomcat and apache.
Is management that gullible that they can be hoodwinked by someone who isn't a sysadmin, but wants to ride on the linux train?