The one thing that generally makes commercial UNIX systems usable are GNU/free/OSS utilties... but basically even before the whole fiasco no one really cared about keeping the SCO ports alive. Even./configure won't save you're bacon any more when you are trying to say, compile Net-SNMP for SCO OpenServer.
The situation as you can imagine is getting worse and there is no sign that certain machines I adminster are going anywhere anytime soon.
Beleive my, compared to your average Linux distro or even Solaris, SCO OpenServer is a pointlessly inscrutable pile of shite.
Voyager was the only decent Star Trek show ever made. Which isn't saying much at all in my book, but it is saying something.
Voyager was aptly named, to me it was like a dungeon crawl set in space. They kept picking up bad-ass new party members, new weapons, better technology, etc. It was like a raiding party... my favorite eps were the ones where they meet new alien race or whatever, try to make friends, they try to steal voyager, voyager and crew whip their ass, take there technology and bail.
I've said this before but the final episodes of voyager should have had them finally getting stupid-warp technology and arriving just in time to find the earth beseiged by well, something bad with the fleet nearly wiped out. Voyager now looks totally different bristling with weapons, maybe they've captured another ship or two. Voyager activates all their stolen borg/whoist weapons and blows them all up real good. Ok, that's totally not how a ST show goes, it's one ship, ostensibly not a warship, going exploring.
Maybe that's why it wasn't so popular, it was stuck between trying to do something different and sticking to the same boring ST themes, plots, etc.
God I feel like a troll but I just can't stand Star Trek or the fans that whine about how great TNG was... especially since I hate that one more than the rest combined.
While it's true I usually spend quite a little while setting up a (desktop) linux system, once it's set up it generally stays that way.
That is, if there are bugs or annoyances that need fixing they stay in that state until I get around to looking at them. I spent 2 weeks off over the holidays with 3 linux and 1 openbsd machine in my house and never tweaked a single one.
As was noted, this isn't my experience with windows, stuff breaks at random and fixing it is often very time consuming. Usually involving trying various combinations of something guessing at the meaning of obscure or non-existant error messages, etc.
Which isn't to say there weren't any components of any of the systems which couldn't benefit from some TLC, but nothing serious.
However, I was just called by my wife's work and they want me to come in and fix their windows printer problems for them. Wonder what I'll get for doing that?
Well, whatever, my post is pointless, however I would like to note that giving free tech support is just painfully stupid, unless it's your boss in which case I suggest you go ahead and do that.
Someone else just dropped off their computer recently for a fix. I had visions of re-installing windows (and with all the drivers that is hours of work.) Instead the hard drive is just dead, so I'm off the hook until they buy a new one. Hopefully they'll just buy a new machine.
MythTV (www.mythtv.org) often figures out where the commercials are all on it's own, you don't even have to skip them yourself.
I no longer know what movies are out, what new shows are coming out, anything that is usually communicated via TV commericals.
Tivo/replaytv might have an edge in reliability, ease of configuration, etc. but MythTV creams them in feature set, hands down. No monthly fees either.
H.323 is obsolete? News to the telco industry. Of course, a lot of stuff is news to the telco industry.
Also, SIP has many problems with NAT and firewalls too. So many that companys make dedicated SIP fixers that attempt to repair SIP packets that have obviously been NATed at one point.
How does NAT obsolete another protocol just because it needs protocol specific hacks? Sounds to me like NAT complicates/breaks any protocol that doesn't involve exactly one connection between two hosts and never references an IP other than in the IP header. To me, that sounds like NAT is what's broken, not SIP/H.323/FTP. Ok, FTP is stupid, but the point stands.
So are people who work at computer stores. If I had a dollar for every time I heard a sales guy at Best Buy lying or making up crap to some poor customer who is obviously relying on him for advice, I'd be rich. Or at least, less poor.
Anyway, sales people talk up what they are selling. We all know that. This applies to game reviews, car salesman, audiophile reviewers, etc.
What I don't understand is why audiophiles get such venom from the slashdot crowd. Many audiophiles feel that there is some (much even) value from subjective listening tests. Many here seem to disagree. But Stereophile also conducts a SERIOUS battery of objective tests, which may or may not make it to the online version (I don't know.)
Is it the terminology used? The terminology represents the subjective nature of the listening tests and the reviewer's enthusiasm for audio and audio gear.
The more I think about the more annoyed I get. No amount of numbers will tell you how something sounds. And no one would reccomend you buy audio gear without listening to it. So the audiophile reviewers are listening to it and telling you how they think it sounds. Perhaps you find that useful, perhaps not. WTF is wrong with that?
I agree... although SpamAssassin has been far more effective than black lists for me I don't like the fact that I now find it easier to just quietly accept all the crap the spammers send out even if the Assassin sorts it out.
With black lists I can bounce it and say ha! I don't want your crap!
Still, I used osirusoft.com explictly and as a part of SpamAssassin (along with other blacklists) and I found it to be a useful resource.
I've never been blacklisted though, so maybe my opinion would change if I had been.
You forget kickbacks and the like. Especially for big purchases there is a lot of wining and dining and ass-kissing and probably (how would I know) some honest to god kickback/bribes.
I wouldn't know, I just implement what some technically unqualified wonk decided we needed often without consultation.
I'm wondering what to do about the 12 month EOL also... I'm trying convince a developer that doesn't even like UNIX (he's a VMS fan) that Linux is the real thing. We use RHL 7.3 in production now... and he's not interested in upgrading at all.
However, if RH stops doing the work of updating packages for me my job gets much much harder. Sure I can track and update all the packages myself. I can even build RPM's moderately proficiently. But I think of the time I'll spend trying to keep RH7.3 updated after it's EOL, and I cringe.
I had a hardware based accelerator for my C64 floppy drive.... it was called Zip something. Anyway, you replaced a chip in the floppy drive or the commodore (or both?) and then you had drilled a hole in the back of the C64 for a switch to enable/disable the accelerated loading. It was orders of magnitude faster when enabled...
Then my brother fried his C64 with a bad power supply. So he grabbed my C64 and fried it too!
If the money were there to port the software to Linux it would probably be under way. As it is we're about to install it on 16 new machines... a sad day that we are lining the pockets of evil.
No one here thinks OpenServer has any kind of future it's just history and timing that keep us on their wagon.
Ok, not a very articulate intro but the point stands.
Great things about MythTV:
number of shows recorded is limited only by the number of tuner cards and how fast your hardware is
can be used with almost ANY video source, and can handle more than one source, like cable+direcTV+antenna
frontend/backend architecture allows you to set up frontends everywhere you have a TV
no monthly fees beyond your internet and cable fees
can play back almost any video format, including DVD's, divx, etc.
can act as mp3 jukebox
will show you the weather on demand
frontend to MAME/MESS/etc.
auto commercial skip (YMMV, it works very well on stations that leave a nice blank frame at the boundaries between the start and end of commercial breaks)
I'm not saying MythTV is perfect or doesn't have drawbacks but in terms of features it makes the other options look decidedly toylike IMHO.
Some things it doesn't do:
collect information about your viewing habbits and recommend shows (I guess people want this, my MythTV machine is full of shows I wanted to watch with little room for anything else)
I'm sure there are others but I can't think of any.
Anyway, not to start a flame war, I'm just saying I like MythTV a lot... and to me it seems superior to TiVo.
I also worked in a NOC that faced serious password overload issues. Now I work in the engineering group at that same company as a UNIX admin and it's almost as bad for me personally as it is for the NOC.
The problem is simply that there is no solution that will do it all. There are simply too many devices in a modern network (especially if your company is into telephony as well) to find an easy solution.
However, the best solution is probably to figure out what the "big three" devices/operating systems are in your network. Then I'd try and figure out what authentication method is supported natively (or with little trouble) by all three. You're in luck with Solaris and most linux distros (all?) because they natively support PAM. (Many other unices probably could use PAM with a little work or may support it out of the box these days.) PAM is nice in that it supports many authentication methods/protocols including LDAP, Radius, and authentication against a mysql database. Cisco supports radius (and probably tacacs, maybe others.) These are the big three I have to deal with and I will probably try to implement a radius solution soon for that reason. But then I also have SCO OpenServer to deal with (unfortunately). And OpenBSD, and HP-UX. And a couple of NT boxes. And some Radlinx PassAPort terminal servers. And some web-servers that need authentication. And likely a whole bunch of other systems/servers/services that I can't remember that could really use to be unified.
The solution? Like the guy above said, stick the info into a database (mysql is painfully easy to set up but obviously you need to pay attention to the security of your mysql server, but you can use Persistent::File perl modules or anything really) and provision radius or kerberos or LDAP or/etc/passwd files from that with perl or the language of your choice. Why put the passwords into a database and not straight into LDAP? For all the reasons that people choose to put data into an sql database usually. (And because then you can claim to be a database admin when you go looking for your next job, and DB's can make some serious money.)
Slashdot Mirror
If only ethereal could display SS7 or ISDN messaging to. If it didn't do H.323 and SIP, my job would be hell.
I find it hard to believe that someone who actually knew perl would go back to using awk.
Depending on how you use perl, it can look as much like C as awk does.
I can't imagine using awk for anything any more, which isn't to say I didn't write thousands of lines of awk code (which in awk, is a lot.)
When you click on next on the first page it takes you back to the angelfire original.
It's not just FSF that has problems with this...
1 07 696705911864&w=2
OpenBSD project isn't including code with the new license either:
http://marc.theaimsgroup.com/?l=openbsd-misc&m=
You, my good man, are a troll.
I administer SCO OpenServer 5.0.5.
./configure won't save you're bacon any more when you are trying to say, compile Net-SNMP for SCO OpenServer.
The one thing that generally makes commercial UNIX systems usable are GNU/free/OSS utilties... but basically even before the whole fiasco no one really cared about keeping the SCO ports alive. Even
The situation as you can imagine is getting worse and there is no sign that certain machines I adminster are going anywhere anytime soon.
Beleive my, compared to your average Linux distro or even Solaris, SCO OpenServer is a pointlessly inscrutable pile of shite.
When I play Axis & Allies as the Japanese, I always attack Russia. It's your only hope of getting enough IPU's to attack the US.
Voyager was the only decent Star Trek show ever made. Which isn't saying much at all in my book, but it is saying something.
Voyager was aptly named, to me it was like a dungeon crawl set in space. They kept picking up bad-ass new party members, new weapons, better technology, etc. It was like a raiding party... my favorite eps were the ones where they meet new alien race or whatever, try to make friends, they try to steal voyager, voyager and crew whip their ass, take there technology and bail.
I've said this before but the final episodes of voyager should have had them finally getting stupid-warp technology and arriving just in time to find the earth beseiged by well, something bad with the fleet nearly wiped out. Voyager now looks totally different bristling with weapons, maybe they've captured another ship or two. Voyager activates all their stolen borg/whoist weapons and blows them all up real good. Ok, that's totally not how a ST show goes, it's one ship, ostensibly not a warship, going exploring.
Maybe that's why it wasn't so popular, it was stuck between trying to do something different and sticking to the same boring ST themes, plots, etc.
God I feel like a troll but I just can't stand Star Trek or the fans that whine about how great TNG was... especially since I hate that one more than the rest combined.
While it's true I usually spend quite a little while setting up a (desktop) linux system, once it's set up it generally stays that way.
That is, if there are bugs or annoyances that need fixing they stay in that state until I get around to looking at them. I spent 2 weeks off over the holidays with 3 linux and 1 openbsd machine in my house and never tweaked a single one.
As was noted, this isn't my experience with windows, stuff breaks at random and fixing it is often very time consuming. Usually involving trying various combinations of something guessing at the meaning of obscure or non-existant error messages, etc.
Which isn't to say there weren't any components of any of the systems which couldn't benefit from some TLC, but nothing serious.
However, I was just called by my wife's work and they want me to come in and fix their windows printer problems for them. Wonder what I'll get for doing that?
Well, whatever, my post is pointless, however I would like to note that giving free tech support is just painfully stupid, unless it's your boss in which case I suggest you go ahead and do that.
Someone else just dropped off their computer recently for a fix. I had visions of re-installing windows (and with all the drivers that is hours of work.) Instead the hard drive is just dead, so I'm off the hook until they buy a new one. Hopefully they'll just buy a new machine.
MythTV (www.mythtv.org) often figures out where the commercials are all on it's own, you don't even have to skip them yourself.
I no longer know what movies are out, what new shows are coming out, anything that is usually communicated via TV commericals.
Tivo/replaytv might have an edge in reliability, ease of configuration, etc. but MythTV creams them in feature set, hands down. No monthly fees either.
I know a company that used 11.0.0.0/8 as their internal address space.
I guess they were Spinal Tap fans or something...
It goes to 11!
No, they probably were just idiots.
H.323 is obsolete? News to the telco industry. Of course, a lot of stuff is news to the telco industry.
Also, SIP has many problems with NAT and firewalls too. So many that companys make dedicated SIP fixers that attempt to repair SIP packets that have obviously been NATed at one point.
How does NAT obsolete another protocol just because it needs protocol specific hacks? Sounds to me like NAT complicates/breaks any protocol that doesn't involve exactly one connection between two hosts and never references an IP other than in the IP header. To me, that sounds like NAT is what's broken, not SIP/H.323/FTP. Ok, FTP is stupid, but the point stands.
So are people who work at computer stores. If I had a dollar for every time I heard a sales guy at Best Buy lying or making up crap to some poor customer who is obviously relying on him for advice, I'd be rich. Or at least, less poor.
Anyway, sales people talk up what they are selling. We all know that. This applies to game reviews, car salesman, audiophile reviewers, etc.
What I don't understand is why audiophiles get such venom from the slashdot crowd. Many audiophiles feel that there is some (much even) value from subjective listening tests. Many here seem to disagree. But Stereophile also conducts a SERIOUS battery of objective tests, which may or may not make it to the online version (I don't know.)
Is it the terminology used? The terminology represents the subjective nature of the listening tests and the reviewer's enthusiasm for audio and audio gear.
The more I think about the more annoyed I get. No amount of numbers will tell you how something sounds. And no one would reccomend you buy audio gear without listening to it. So the audiophile reviewers are listening to it and telling you how they think it sounds. Perhaps you find that useful, perhaps not. WTF is wrong with that?
Idiot.
The point is that projects like MythTV and Freevo won't be able to get at the data unless someone can implement this "standard."
Get a clue.
I agree... although SpamAssassin has been far more effective than black lists for me I don't like the fact that I now find it easier to just quietly accept all the crap the spammers send out even if the Assassin sorts it out.
With black lists I can bounce it and say ha! I don't want your crap!
Still, I used osirusoft.com explictly and as a part of SpamAssassin (along with other blacklists) and I found it to be a useful resource.
I've never been blacklisted though, so maybe my opinion would change if I had been.
That's very sharp... I'll have to set that up.
I wish I would have noticed sooner... for some reason I seem to have lost all email from every mailing list I'm on, but not much else.
You forget kickbacks and the like. Especially for big purchases there is a lot of wining and dining and ass-kissing and probably (how would I know) some honest to god kickback/bribes.
I wouldn't know, I just implement what some technically unqualified wonk decided we needed often without consultation.
I'm wondering what to do about the 12 month EOL also... I'm trying convince a developer that doesn't even like UNIX (he's a VMS fan) that Linux is the real thing. We use RHL 7.3 in production now... and he's not interested in upgrading at all.
However, if RH stops doing the work of updating packages for me my job gets much much harder. Sure I can track and update all the packages myself. I can even build RPM's moderately proficiently. But I think of the time I'll spend trying to keep RH7.3 updated after it's EOL, and I cringe.
I had a hardware based accelerator for my C64 floppy drive.... it was called Zip something. Anyway, you replaced a chip in the floppy drive or the commodore (or both?) and then you had drilled a hole in the back of the C64 for a switch to enable/disable the accelerated loading. It was orders of magnitude faster when enabled...
Then my brother fried his C64 with a bad power supply. So he grabbed my C64 and fried it too!
We use SCO OpenServer quite extensively too.
If the money were there to port the software to Linux it would probably be under way. As it is we're about to install it on 16 new machines... a sad day that we are lining the pockets of evil.
No one here thinks OpenServer has any kind of future it's just history and timing that keep us on their wagon.
Great things about MythTV:
I'm not saying MythTV is perfect or doesn't have drawbacks but in terms of features it makes the other options look decidedly toylike IMHO.
Some things it doesn't do:
Anyway, not to start a flame war, I'm just saying I like MythTV a lot... and to me it seems superior to TiVo.
http://www.mythtv.org
I have no suggestions that apply to Unixware, if you had some OpenServer systems, I could probably help.
OpenServer 5.0.0 would lock up if you ran nmap -O on it. Or you could just wait for the bugs in the license daemon to crash the machine for you.
Strange, but the fonts in Mozilla and Galeon look the same to me. Am I just tripping?
We throw nerf balls at each other and scream obscenities at customers while the phone is muted.
It seems to help.
Thanks Microsoft! I surely did want your bloated junk running all the time or I wouldn't run your OS in the first place!
I also worked in a NOC that faced serious password overload issues. Now I work in the engineering group at that same company as a UNIX admin and it's almost as bad for me personally as it is for the NOC.
The problem is simply that there is no solution that will do it all. There are simply too many devices in a modern network (especially if your company is into telephony as well) to find an easy solution.
However, the best solution is probably to figure out what the "big three" devices/operating systems are in your network. Then I'd try and figure out what authentication method is supported natively (or with little trouble) by all three. You're in luck with Solaris and most linux distros (all?) because they natively support PAM. (Many other unices probably could use PAM with a little work or may support it out of the box these days.) PAM is nice in that it supports many authentication methods/protocols including LDAP, Radius, and authentication against a mysql database. Cisco supports radius (and probably tacacs, maybe others.) These are the big three I have to deal with and I will probably try to implement a radius solution soon for that reason. But then I also have SCO OpenServer to deal with (unfortunately). And OpenBSD, and HP-UX. And a couple of NT boxes. And some Radlinx PassAPort terminal servers. And some web-servers that need authentication. And likely a whole bunch of other systems/servers/services that I can't remember that could really use to be unified.
The solution? Like the guy above said, stick the info into a database (mysql is painfully easy to set up but obviously you need to pay attention to the security of your mysql server, but you can use Persistent::File perl modules or anything really) and provision radius or kerberos or LDAP or /etc/passwd files from that with perl or the language of your choice. Why put the passwords into a database and not straight into LDAP? For all the reasons that people choose to put data into an sql database usually. (And because then you can claim to be a database admin when you go looking for your next job, and DB's can make some serious money.)
Good luck! (And someone wish me luck too.)