> Anyone can be brilliant. Anyone can be a jerk. Sometimes these two things overlap. I'm not convinced that there's a higher penetration of this in IT than any other profession.
I think it is, for 2 reasons:
1) People (still) don't understand IT, so it's still a priesthood, especially at the upper levels.
2) As several people have noted in books, the span of productivity between mediocre programmers and heroes is larger than in almost any other endeavour; therefore the span of social behaviour which is likely to be tolerated is probably wider than for people who mop floors.
Figher pilots get away with lots more shit than deck-swabs do, too.
And let us not mention those cases where consensual sex gets someone lined up for fraternization charges, and they figure it's better to throw they guy under the bus...
As has been noted here, almost every doctor accepts assignment to some government reimbursement program like Medicare or Medicaid.
Since "[a]lthough the First Amendment only explicitly applies to the Congress, the Supreme Court has interpreted it as applying to the executive and judicial branches. Additionally, in the 20th century the Supreme Court held that the Due Process Clause of the Fourteenth Amendment applies the limitations of the First Amendment to each state, including any local government within a state.".. it's likely that a case could be made that a doctor doing this is violating 42 USC 1983, depriving you of a civil right (due process, if not free speech) "under color of law", and that you could sue them personally for treble damages.
Even when you included the cost of retraining every single employee who had to *work with* that application?
Everyone always seems to forget this one.
Allied Van Lines, about 8 years ago, replaced the 2-screen CICS app called CAMIS with a new thing called AMS that was much fluffier looking and reportedly easier to train.
But it was 7 screens per transaction instead of 2, and required them to retrain all the people they had... and last I looked, *still* wouldn't do all the stuff the old one had.
And they didn't have exceptionally high turnover in the positions that this package supported in the first place.
So: more than tripled the transaction load on the mainframe. Check. Increased the training load on the present employees. Check. Didn't accomplish anything productive. Check.
And on a zSeries, I might actually say "100.00%"; mainframes really are impressive, impressive pieces of machinery.
The canonical story, of course, is the guy who ran 44,763 copies of Linux directly under z/VM (I think it was) before the machine he was on said "Ok, then; that's enough". He wasn't *doing* anything, of course, but let's assume you could only get 5% of that number running real loads on the same hardware.
That's *2 cabinets*. How many 1RU dual quads can you fit in 2 cabinets? 84? So, 672 cores.
Compared to 2,000ish VMs.
And, really: 5 year uptimes, punctuated only by "We need to shut it down because we need to upgrade the shunt bypass on the UPS that feeds it". (True story)
The way I heard the assertion was that when Cutler came over and designed NT, the API looked suspiciously like that of VMS -- to anyone who programmed on VMS.
In point of fact, many debit card issuers now offer the same sort of guarantees -- Visa Check Card branded cards have a national ad campaign touting this specific fact.
I gather the amusingly named RTFM (Request Tracker FAQ Manager) is the solution for the problem you note, though I'm only just 2 weeks into my latest install of 3.8.2, and haven't gone there yet.
RT's not bad, but there are some spots where I find myself tripping over they way it wants to do things: there are are some environments for which it's much better suited than others.
Specifically, I am now in an 'internal IT' environment; my opinion of RT was that it didn't deal all that well with my previous environment, which was 'external paid consultant' -- it doesn't really do "customer file", and the attendent things like "any email tickets from this email address are for customer Y; automatically add email Z (their supervisor) as a watcher". I would *like* that in my current environment; I *needed* it in the other one, and it just won't do it.
I also need something like Asset Tracker, but again, I need it more tightly integrated than AT is -- the ability to pick the most likely assets about which a user might be reporting a problem is pretty important, and I couldn't see how you'd do it from looking at the available pre-install information, which is sort of thin.
The best suggestion anyone had was to treat customers as assets in AT, but that required a level of training for ticket writers that simply wasn't going to fly.
So yeah, maybe RT will suit you, but as I note in the lead to the (still pretty thin) user manual on the wiki: "Don't install it Sunday if you need to write tickets Monday morning".
It is, in the phrasing of... Brooks?... a program. Not a Systems Program Product.
At least, it isn't for free; what you get if you give money denominated in three zeroes to Best Practical, I can't say.
who thinks it's entirely appropriate that this is the module with the UPA and the toilets in it (which is intended to be a loving observation, and not a crack, in case you were wondering)...
Incidentally, in case y'all haven't look *into* said toilets, they have an interesting design.
Near as I can figure, the shit is *supposed* to hit the fan.
(Yes, yes, I stole it from Spider, ok? I'll be here all week.)
Science is, well, it's science. It's the same everywhere.
For many, *their perception* of science is that it's a religion, but all that means is that they have a very loose grasp on the meaning of the word "fact".
Which disqualifies them from being entitled to an opinion.
As Heinlein once said, "Stupid is stupid. Religion doesn't make it smart."
> The question is whether this kind of change and this rapid is a good thing for human interests.
No.
The *question* is whether we had anything to do with it, and whether we *can* have anything to do with it.
> It doesn't help the debate when climate change deniers make a big deal
What *really* doesn't help matters is when the people who support a diagnosis of anthropogenia in global warming, er, um, "climate change" co-opt the argument by subtly characterizing their opponents in the same stratum as people who don't believe Hitler gassed 6 million Jews in WWII.
I hereby propose Ashworth's Corollary to Godwin's Law:
"Utilizing the word 'deniers' to refer to your opponent in an argument triggers Godwin's Law."
That thing is 12 years old now (about as old as my slashdot ID, come to think of it:-), and it crops up at the oddest times.
There are several good references in the actual RFC, which is worth checking out for that reason.
(I post the link to netfunny because they didn't break one of the jokes in a vapid attempt to make the format be correct; Jon accepted that one, but I don't think he edited it.)
Clearly, you people never read Tom Limoncelli (or, well, me).
Never bake functional naming or geography into server names -- if you are *lucky*, your company (or your employment there) will last long enough for you to hate yourself for doing that.
There really is a justifiable reason why server naming should be at least semi-arbitrary.
Certainly, people should be able to use functional names to access boxes. But that's *DNSs* issue, not the hostname's.
This brings up an issue that I asked a friend of mine, a correspondent blogger for Pajamas Media who was credentialled for the DNC in Colorado, to ask if he ever got with anyone influential in the Obama campaign. I don't think he got the chance to ask.
It's the logical outgrowth of "all the smart people don't work for you" and it is, roughly: "What plans does the Obama administration have for Idea Triage from the general public?".
I fully believe the "it's who you know" depiction from fiction of the way that good ideas percolate up to someone who can do something about them in the Federal government, and I don't think it's good enough.
Surely you'll wade through a lot of crap, but just as surely, there has to be a serviceable way to build an organization to which people can submit ideas and people who are sufficiently educated and street smart to spot the good ones can.
Our current community moderation systems are a start, as long as they don't turn into the Tyranny of the Majority; not all good ideas are popular.
Like, oh, I dunno, the US Constitution.
Perhaps I should make this point to Bill Joy today, before he takes the job. Any slashdotters have him on a DSS key on their desk phone?:-)
I'm going to toss an oar in the water because I'm deploying Zimbra for my company to replace (please, put your beverage down) Exchange 5.5 on Win2k with an NT4 PDC.
Ok, now that you're done laughing...
Zimbra is, to use the words Jerry Pournelle once used to describe Vulcan (dBase I), "infuriatingly excellent".
Within the current limts of AJAX, it's web client program is very nice indeed; they have a detachable client but I haven't played with it yet.
The system runs atop a lotta buncha FOSS packages, though it brings them all along with it, which means you really want to dedicate a box/VM to running it -- this is a feature, though, not a bug. Why? Because it means that *they* worry about upstream security bugs, not you.
It does POP/SSL and IMAP/SSL, and the webclient itself can be locked to only run SSL, if you like. It has a very nice multi-domain admin control panel, the commercial version will do hot backups and connect to Outlook, and there's a Migration Wizard to pull mail, contacts and calendars out of Exchange Server.
That said, we now proceed to the infuriating part.
There are lots of things that I (having been a mail admin for 10 or 15 years, and moving about 500 real messages a day over 15 mailing lists) think it ought to do that it doesn't.
The two most fundamental are that it doesn't thread on In-Reply-To but on message title, and that it doesn't handle mailing list traffic too well. The former is Broken As Designed: there has been a bug on their (open) Bugzilla about this since v3.mumble; they just shipped 5.0.10, but no progress on the bug, no official comment, and they decline to *close* the damned bug as well -- I think that this falls in the category of "keeping all your nuts in one basket".
On the latter front, there are "next unread" and "previous unread" keys, but since they paginate their message list (for reasons that I publically assume have to do with shitty AJAX toolkits and no one disagrees with me), it would be nice, you'd think, if those went *over* the edges of pages; they don't. Since that's true, you have to read your mailing list mail backwards -- since there's no practical way to get to the *beginning* of the new traffic in the folder if you sort forwards.
There are other foibles, but perhaps business (and *maybe* college) users wouldn't notice them; they're largely the result of growing up on Mutt.
Mutt definitely sucks less than Zimbra; I haven't filed 36 bugs on Mutt.
Go into it with your eyes open, certainly, but for all that I'm personally annoyed with it, Zimbra has some good things to recommend it.
John Holder, Mike Morse, and a couple of the other staffers who frequent their forum are pretty good guys.
And in the last month, denizens thereof have rolled out 22K and 47.5K mailbox installs. So clearly it scales. Will you have to learn some things? Yes. Well it be perfect, and roll out to a college sized install with no problems whatever? Well, maybe, but I'd plan for a *few* annoyances.
Should you ignore it?
Only at your peril.
Outside hosting is, as has already been noted, extremely fraught with legal landmines.
And this month has, I think, displayed quite nicely the risks of failing to heed warnings.
Slashdot Mirror
> Anyone can be brilliant. Anyone can be a jerk. Sometimes these two things overlap. I'm not convinced that there's a higher penetration of this in IT than any other profession.
I think it is, for 2 reasons:
1) People (still) don't understand IT, so it's still a priesthood, especially at the upper levels.
2) As several people have noted in books, the span of productivity between mediocre programmers and heroes is larger than in almost any other endeavour; therefore the span of social behaviour which is likely to be tolerated is probably wider than for people who mop floors.
Figher pilots get away with lots more shit than deck-swabs do, too.
For us poor USAdians, that joke isn't especially funny, since we write it 7/22.
Indeed.
That's why my preferred neologism is "USAdians", which is -- at the very least -- pronounceable.
And, in my experience, most of the people using things like USAdians are being ironic; picking on people who really *are* xenophobes.
And let us not mention those cases where consensual sex gets someone lined up for fraternization charges, and they figure it's better to throw they guy under the bus...
No it's not.
Anything over about 60ms is unacceptable; 40's better.
My Olympus E-10 manages 60 reliably, and it's 9 frickin years old.
Yes, but does he leave you open on the table and go to the pub for lunch and a pint?
(When he finally gets to you, 6 months late)
As has been noted here, almost every doctor accepts assignment to some government reimbursement program like Medicare or Medicaid.
Since "[a]lthough the First Amendment only explicitly applies to the Congress, the Supreme Court has interpreted it as applying to the executive and judicial branches. Additionally, in the 20th century the Supreme Court held that the Due Process Clause of the Fourteenth Amendment applies the limitations of the First Amendment to each state, including any local government within a state.".. it's likely that a case could be made that a doctor doing this is violating 42 USC 1983, depriving you of a civil right (due process, if not free speech) "under color of law", and that you could sue them personally for treble damages.
1983 is a *really cool* statute. Look it up.
http://www.winhistory.de/more/386/xpmini_eng.htm
> Checkout the x-fry/x-bender headers: echo -e "HEAD / HTTP/1.1\nHost: slashdot.org\n\n" | netcat slashdot.org 80
Isn't wget -S easier?
Even when you included the cost of retraining every single employee who had to *work with* that application?
Everyone always seems to forget this one.
Allied Van Lines, about 8 years ago, replaced the 2-screen CICS app called CAMIS with a new thing called AMS that was much fluffier looking and reportedly easier to train.
But it was 7 screens per transaction instead of 2, and required them to retrain all the people they had... and last I looked, *still* wouldn't do all the stuff the old one had.
And they didn't have exceptionally high turnover in the positions that this package supported in the first place.
So: more than tripled the transaction load on the mainframe. Check. Increased the training load on the present employees. Check. Didn't accomplish anything productive. Check.
Kept the corporate IT budget in good shape.
Check.
Nope. There are probably half a dozen botnet operators about whom that description would be accurate, as well.
I believe you've mispelt "100.0%".
And on a zSeries, I might actually say "100.00%"; mainframes really are impressive, impressive pieces of machinery.
The canonical story, of course, is the guy who ran 44,763 copies of Linux directly under z/VM (I think it was) before the machine he was on said "Ok, then; that's enough". He wasn't *doing* anything, of course, but let's assume you could only get 5% of that number running real loads on the same hardware.
That's *2 cabinets*. How many 1RU dual quads can you fit in 2 cabinets? 84? So, 672 cores.
Compared to 2,000ish VMs.
And, really: 5 year uptimes, punctuated only by "We need to shut it down because we need to upgrade the shunt bypass on the UPS that feeds it". (True story)
Google Linux+390 and read a little bit...
The way I heard the assertion was that when Cutler came over and designed NT, the API looked suspiciously like that of VMS -- to anyone who programmed on VMS.
You youngsters with 6 digit IDs...
WinAMP notwithstanding, that's "cue up some Elton John".
Just ask Johnny Fever; he'll tell you...
In point of fact, many debit card issuers now offer the same sort of guarantees -- Visa Check Card branded cards have a national ad campaign touting this specific fact.
I gather the amusingly named RTFM (Request Tracker FAQ Manager) is the solution for the problem you note, though I'm only just 2 weeks into my latest install of 3.8.2, and haven't gone there yet.
RT's not bad, but there are some spots where I find myself tripping over they way it wants to do things: there are are some environments for which it's much better suited than others.
Specifically, I am now in an 'internal IT' environment; my opinion of RT was that it didn't deal all that well with my previous environment, which was 'external paid consultant' -- it doesn't really do "customer file", and the attendent things like "any email tickets from this email address are for customer Y; automatically add email Z (their supervisor) as a watcher". I would *like* that in my current environment; I *needed* it in the other one, and it just won't do it.
I also need something like Asset Tracker, but again, I need it more tightly integrated than AT is -- the ability to pick the most likely assets about which a user might be reporting a problem is pretty important, and I couldn't see how you'd do it from looking at the available pre-install information, which is sort of thin.
The best suggestion anyone had was to treat customers as assets in AT, but that required a level of training for ticket writers that simply wasn't going to fly.
So yeah, maybe RT will suit you, but as I note in the lead to the (still pretty thin) user manual on the wiki: "Don't install it Sunday if you need to write tickets Monday morning".
It is, in the phrasing of... Brooks?... a program. Not a Systems Program Product.
At least, it isn't for free; what you get if you give money denominated in three zeroes to Best Practical, I can't say.
who thinks it's entirely appropriate that this is the module with the UPA and the toilets in it (which is intended to be a loving observation, and not a crack, in case you were wondering)...
Incidentally, in case y'all haven't look *into* said toilets, they have an interesting design.
Near as I can figure, the shit is *supposed* to hit the fan.
(Yes, yes, I stole it from Spider, ok? I'll be here all week.)
Science is, well, it's science. It's the same everywhere.
For many, *their perception* of science is that it's a religion, but all that means is that they have a very loose grasp on the meaning of the word "fact".
Which disqualifies them from being entitled to an opinion.
As Heinlein once said, "Stupid is stupid. Religion doesn't make it smart."
> The question is whether this kind of change and this rapid is a good thing for human interests.
No.
The *question* is whether we had anything to do with it, and whether we *can* have anything to do with it.
> It doesn't help the debate when climate change deniers make a big deal
What *really* doesn't help matters is when the people who support a diagnosis of anthropogenia in global warming, er, um, "climate change" co-opt the argument by subtly characterizing their opponents in the same stratum as people who don't believe Hitler gassed 6 million Jews in WWII.
I hereby propose Ashworth's Corollary to Godwin's Law:
"Utilizing the word 'deniers' to refer to your opponent in an argument triggers Godwin's Law."
I was thinking Crazy Pierre, myself.
You like me.
You really like me. :-)
That thing is 12 years old now (about as old as my slashdot ID, come to think of it :-), and it crops up at the oddest times.
There are several good references in the actual RFC, which is worth checking out for that reason.
(I post the link to netfunny because they didn't break one of the jokes in a vapid attempt to make the format be correct; Jon accepted that one, but I don't think he edited it.)
Clearly, you people never read Tom Limoncelli (or, well, me).
Never bake functional naming or geography into server names -- if you are *lucky*, your company (or your employment there) will last long enough for you to hate yourself for doing that.
There really is a justifiable reason why server naming should be at least semi-arbitrary.
Certainly, people should be able to use functional names to access boxes. But that's *DNSs* issue, not the hostname's.
This brings up an issue that I asked a friend of mine, a correspondent blogger for Pajamas Media who was credentialled for the DNC in Colorado, to ask if he ever got with anyone influential in the Obama campaign. I don't think he got the chance to ask.
It's the logical outgrowth of "all the smart people don't work for you" and it is, roughly: "What plans does the Obama administration have for Idea Triage from the general public?".
I fully believe the "it's who you know" depiction from fiction of the way that good ideas percolate up to someone who can do something about them in the Federal government, and I don't think it's good enough.
Surely you'll wade through a lot of crap, but just as surely, there has to be a serviceable way to build an organization to which people can submit ideas and people who are sufficiently educated and street smart to spot the good ones can.
Our current community moderation systems are a start, as long as they don't turn into the Tyranny of the Majority; not all good ideas are popular.
Like, oh, I dunno, the US Constitution.
Perhaps I should make this point to Bill Joy today, before he takes the job. Any slashdotters have him on a DSS key on their desk phone? :-)
I'm going to toss an oar in the water because I'm deploying Zimbra for my company to replace (please, put your beverage down) Exchange 5.5 on Win2k with an NT4 PDC.
Ok, now that you're done laughing...
Zimbra is, to use the words Jerry Pournelle once used to describe Vulcan (dBase I), "infuriatingly excellent".
Within the current limts of AJAX, it's web client program is very nice indeed; they have a detachable client but I haven't played with it yet.
The system runs atop a lotta buncha FOSS packages, though it brings them all along with it, which means you really want to dedicate a box/VM to running it -- this is a feature, though, not a bug. Why? Because it means that *they* worry about upstream security bugs, not you.
It does POP/SSL and IMAP/SSL, and the webclient itself can be locked to only run SSL, if you like. It has a very nice multi-domain admin control panel, the commercial version will do hot backups and connect to Outlook, and there's a Migration Wizard to pull mail, contacts and calendars out of Exchange Server.
That said, we now proceed to the infuriating part.
There are lots of things that I (having been a mail admin for 10 or 15 years, and moving about 500 real messages a day over 15 mailing lists) think it ought to do that it doesn't.
The two most fundamental are that it doesn't thread on In-Reply-To but on message title, and that it doesn't handle mailing list traffic too well. The former is Broken As Designed: there has been a bug on their (open) Bugzilla about this since v3.mumble; they just shipped 5.0.10, but no progress on the bug, no official comment, and they decline to *close* the damned bug as well -- I think that this falls in the category of "keeping all your nuts in one basket".
On the latter front, there are "next unread" and "previous unread" keys, but since they paginate their message list (for reasons that I publically assume have to do with shitty AJAX toolkits and no one disagrees with me), it would be nice, you'd think, if those went *over* the edges of pages; they don't. Since that's true, you have to read your mailing list mail backwards -- since there's no practical way to get to the *beginning* of the new traffic in the folder if you sort forwards.
There are other foibles, but perhaps business (and *maybe* college) users wouldn't notice them; they're largely the result of growing up on Mutt.
Mutt definitely sucks less than Zimbra; I haven't filed 36 bugs on Mutt.
Go into it with your eyes open, certainly, but for all that I'm personally annoyed with it, Zimbra has some good things to recommend it.
John Holder, Mike Morse, and a couple of the other staffers who frequent their forum are pretty good guys.
And in the last month, denizens thereof have rolled out 22K and 47.5K mailbox installs. So clearly it scales. Will you have to learn some things? Yes. Well it be perfect, and roll out to a college sized install with no problems whatever? Well, maybe, but I'd plan for a *few* annoyances.
Should you ignore it?
Only at your peril.
Outside hosting is, as has already been noted, extremely fraught with legal landmines.
And this month has, I think, displayed quite nicely the risks of failing to heed warnings.
That joke went out with modem noises on radio and TV commercials.
Oh, wait...