Yes, but when I was in high school back in 1989 we had 1 classroom with PC's and a 4 of 5 in the library. So the student to computer ratio was about 1:100.
Now they want a computer in every classroom for every student. And then one on every teachers desk, and another high end lab in the library. The student to computer ratio is now approaching 2:1 in favor of the computers.
And adjusted for inflation, the cost of managing 25 standalone pc's running DOS and a copy of some math program that resides on a pair of floppies is nothing compared to the cost of 25 pc's running windows 7, connected to the internet, with few dozen apps.
Computers are great and all, but I have yet to see a good study that showed that the kids with ubiquitous computer access were better off than those who had limited computer access.
Start with an iPhone 4s or Android equiv (http://www.pcworld.com/article/241955-2/smartphone_camera_battle_iphone_4s_vs_the_android_elite.html)
Once you run into the limits of what that platform can do, then get a better camera. But by then you will have a better idea of why the phone does not work for you. If it is focus speed, go high end DSLR; if it is image quality, go 4/3rds or low end DSLR, etc.
For my chosen field (sports photography), a Nikon D300+MB10D w/ 70-200mm/2.8, 17-55/2.8 and SB-900 is entry level (sports photography) but comes at a high cost (as in $4-5k). For others, maybe a mirrorless system (4/3rds, Nikon 1, etc) would be a better option and a full system would cost less than one of my lenses.
Don't take personal offense but step back and look at the client -> agency -> contractor relationship and how fundamentally screwed up it is.
Most importantly, ask yourself if your primary purpose in being at a clients site is to make the customer happy or if it is to make your employer money.
We all want to say our primary job it is to make the customer happy, but in reality our primary job is to make the employer employer money (or else you wont be in business very long. Customers are usually happier if you don't charge them).
Now are you doing your employer (the folks who sign your checks) any favors by reducing the long term amount of work that needs to be done?
You may say counter by saying that if you do a good job then the customer will hire you back and it makes your employer more money in the long run. In reality, if the customer was qualified to determine if you did a good job (fix it permanently) vs an adequate one (patch it), then you would probably not be there to being with (otherwise they would do it themselves). As others have said in the past, "adequacy is sufficient, all else is superfluous."
Yes, this is a very Machiavellian view of the contracting world.
This has nothing to do with how you as an individual approach a given situation. You may have a higher personal ethical standard and want to go well above and beyond the call of duty. I've met a number of consultants/contractors/perms/temps/whatever who are like this. I like working with these people, mostly because they will work as a team and do what is right for the project as a whole and not just their department or group.
But I also have to face the reality that the contracting environment as it is setup today is fundamentally flawed in that it financially encourages mediocre work.
On average I'd say 2/3rd fail to make it past the first month. We have had some that didn't even make it past the first week. (HINT: do not make sexist comments to your female coworker. )
We don't try and trap them, but it seems that their inflated sense of self worth and job skills are such that we think that they actually go out of their way to screw up and then argue with us about how they didn't screw up. We typically give them a 2nd chance to fix it once we point it out to them. The thought is that if they can learn from their mistakes then they have hope. If not, well.... Next...
I see reading comprehension is not one of your strong points, but I'm happy to say that you are very good at jumping to conclusions.
As I said before, as a fed contractor, I was restricted to working more than that. I could want to have be there for 80, but the higher ups told me to not screw up the accounting, and not to even think about going off clock since an audit would nail them. When I left they back fill me with two new folks. My former co-workers bitched about my leaving, but the 50% pay raise was too much to ignore and I was bored out of my mind there.
So I jumped to a consulting firm in the middle of the dot com boom days. In one year I billed almost 3000 hours. Take out the vacation and 2 weeks from having a kid and I'm over 70 as an average. Add in the crap that was off the clock and it is higher.
If you have been around for 20+ years you should know that on average you get most of your work done in the first 20 hours of the work week. After that its an exponential decay. After 60 you are just taking up space and consuming coffee. (The occasional crunch time is one thing, but I'm talking averages). But when you are a consultant billing $300/hr, the company who is cutting your checks really does not care as long as the customer is happy.
So much for your assessment of my abilities.
And the odd thing is that the group that I now work with in a local county government (all direct) is by far the best damn group I have ever seen or had to work with. Based on your off the cuff assessment, we should not exist. We should all want to leave for more money and the thrill of working long hours.
What you fail to understand is that it not the type of job you are in. I've seen good and bad groups of contractors, consultants, government workers, education, commercial, etc. I've run into incredible people in places I least expected to. I've also run into some real idiots in spots where you wonder how they ever survived there. You name an industry or group and I'm sure somebody will toss out an example of a good team and an example of a crap team. In the end it is the people you work for and with. Good people want to work with other good people. Its a positive feedback cycle. A good group knows when they are talking to other good people. The flip side is also true. Bad people work with other bad people (and the dunning-kruger effect is in full effect in those environments)
And please don't confuse productivity and quality with 60+ hour work weeks, or even 45+. We had a security guy who did 70+ every week. He was a hard worker and knew the talk but fundamentally lacked any real skills. When he left to become a consultant we replaced him with a few small shell scripts. (he spent 20+ hours just creating new user accounts; by hand in ldap. One guy wrote a script and it now takes us 10 minutes a week, if that) The firm he now works for has no clue that he has no clue, because the entire company has no clue. But the owner knows some folks with three letter titles from the golf course and they get security assessment gigs because of it. C'est la vie.
Disclamer: I was a salaried contractor who worked for Northrop who was sub for Lockhead who worked for the USAF. A typical oversized, multi-year, multi-billion dollar contract.
Northrop could not replace me at any time they wanted without cause any easier than my current government employer could. (my current employer is a local county.) Nobody in that entire building ever got fired "just because." Its often bantered about, but given that most of the contractors were ex-military in their 40's (I was one of two exceptions) it never happened. Once they got you in there, it was simply too expensive to swap you unless you royally screwed up and the customer wanted you out of there. And I got full benefits from Northrop, including vacation, sick, medical, 401k, etc, etc. I get more or less the exact same benefits from my current employer (401k, no pension).
I was salaried, but I had to declare all of my hours to the project for billing purposes. They did not want me to burn hours faster than what was allotted in the contract, so they really didn't want me there more than 40 unless something major came up. My current job is also salaried, and on average I probably work the same average number of hours.
Now, individual situations can alter this perception. Maybe you worked for SAIC or some other warm body shop in a contract from hell that had high turnover. But in all of those cases, the issue was not that folks got fired, it's that they all left the second they found a better paying job because the contract went to the lowest bidder, and SAIC only wanted to pay 1/3rd to 1/4 of the rate to the actual person doing the job to cover their "overhead" ($60/hr for a system admin to SAIC results in $20/hr to the actual contractor. Any takers for a unix admin with a security clearance for @$30-40k/yr? No? Didn't think so.).
We do use contractors/consultants at my current job in two cases:
1) Very short term 'in, do xyz, get out' implementation or staff augmentation projects. Basically, install this software, train the IT staff and get out. Those are all $250+/hr consultants and we limit how long they are in house. They are very expensive ($500k/yr) but we only use them to get over critical staff shortages or crunch periods.
2) Contract to hire for new folks at the help desk. Mostly because we can, er, fire them faster. That said we have yet to do it "just because." In all cases the idiot screwed up not once, not twice but three times within the first two weeks and got walked. But when you get here those conditions are very explicit. Going through the interview phase is a pain in the ass for us (we'd rather be doing real work) so once somebody is here we'd rather it work out than not. It actually cost us more to go that route but the process is faster so nobody cares.
Having worked in education, dot.com startups (during the boom and bust years), federal (DoD) and now Local government, I have found that the desire to "do things properly'"is not a trait of the type of job (contract, perm, etc), it is a trait of the individual.
Sorry, I've worked both sides of this fence, and you should have stopped before you even typed the first word.
The contractors for the fed/military/etc do not work day to day, twice the hours or have triple the productivity.
They are given year+ long contracts, work the same hours, and have the same or less productivity. The perms face the exact same thing, their entire division can be wiped away with the stroke of a budgeting pen.
We are not talking about day labors here, all federal contracts are long and well defined. While your project may get canceled with the next _YEARLY_ budget, the odds of it suddenly going under are next to 0.
As a fed contractor, I never put in more than 40 hours a week. That is what we had in the budget, and to do more than that would have resulted in issues. The "cost+" contracts that would let me work 80 hour weeks and have the contracting agency get paid for it are few and far in between. Most are fixed at the rates and the number of hours, it does them no good to have you work more than your scheduled rate.
The productivity thing is pure bullshit. I've seen incompetent admins on both sides, but most are on the contracting side because the contracting firm wants to keep a larger % of the cut to themselves, and thus toss inexperienced newbies into the slot in the hopes that nobody will notice. The real kicker is that as a contractor you have an incentive to not really fix things, but to just patch them. After all, why fix something once and for all when your job depends on the customer needing to have you around to constantly fix something?
But its not just by zip: its zip, street and address. Tax districts can vary house by house. 100 Main St. can be in a different tax district than 101 Main St. We have a road named, aptly enough, 'County Line Road'. Homes on one side are in one county, homes on the other side a different county. Both use the same zip code because the post office does not care.
Now as a business you can use a simplified zip = X% mapping but if you under collect then you are screwed. (If you over collect nobody cares).
Back in the day companies could get away a simple lookup table and hope for the best simply because nobody was really looking. We (fyi: I work for a county) now have an app that integrates with our gis parcel data and geocodes the entire thing and then sends the State a bill for what they owe us based on what they collected. In its first month of use the app has 'saved' us about a 100k in what was unclaimed funds because in the past we were just keying off of the zip code.
There is, but it involves geocoding every single address. And then updating it every time any one of the 60,000 tax districts change their boundaries or rates.
Here is the problem, you can have two houses on opposite sides of the street be in two different tax districts. So a simple 'if zip == xxxxx, then tax = Y' type of lookup table will not work.
You then have the issue of the corporation needing to potentially apply for a sales tax license in jurisdiction before they can collect the tax.
Then you have the issue of having to possibly send the check to 3 or 4 different groups on different schedules for each customer in a different.
And finally there is the question of what gets taxed. In some states, some items are not taxed (usually basic food). So if I order a 10lb tub of powdered gatorade from amazon.com it may get taxed in one state but not another, both of which have a sales tax.
To call it a mess is an understatement. This is the main reason why the courts tossed out the states requirement to collect the tax: the burden was simply too much. If memory serves me correctly, that same court decision left the door open to enact a simplified sales tax scheme (if shipping to NY, then charge X% and send it to Y address and be done with it).
I picked up an HP LaserJet 4100 (with duplexer and jetDirect card) for $25 at a used pc sale run by the local county government. I also got an HP 8150DN (duplexer, network and 2000 sheet feeder tray) at the same sale for another $25.
The things are built well and everybody supports them. Because they were so common, toner is easy to find. Not that I'll need it, the 8150 came with two full cartridges rated at 20,000 sheets each.
Oh, and nobody goes on holiday without contact for over 24 hours, do they? I bring a laptop and a smartphone with me wherever I go. Even when I visited Northern Africa, I made sure to get online at least once a day to check, act on, and reply to my email.
Its not a vacation if you can find me.
I leave my cell, laptop, etc home. For my last trip, I told my co-workers what park I would be in and that if something went south that they can call the park ranger and then hope that they can find me.
I want to get away from the the regular grind, not bring them with me:-)
Technically, at this time oracle does not own sun.
They have announced that they will purchase them and the sale is pending, but until that time the two companies are totally independent and functionally must continue to operate as such.
So sometime this summer the oracle logo will be correct, but currently it is wrong.
Cisco + Sun would make more sense. Mostly because there is very little overlap in their actual products but their two lines constantly need to work together. (Our sun servers are connected to Cisco ethernet switches, our SunRays vpn into Cisco vpn concentrators, our Sun Storage is connected to Cisco MDS switches, etc). It would also give Cisco the biggest, baddest InfiniBand switch on the market (and at 110Tbps, its switching capacity totally trashes anything cisco has ever produced).
The biggest problem with the Sun+IBM deal was that there was so much overlap, customers would be left to wonder which product lines would get discontinued. (glassfish vs websphere, solaris vs aix, sparc vs power, sun's servers vs ibm's, storage, tape, etc, etc, etc. )
I never touched a computer till the 7th grade, and never did anything more than basic word processing with them till the 10th grade. I learned how to do calculus using chaulkboards and paper.
My kid is in the 2nd grade and already doing powerpoint. WTF?!?!?! The focus is on presentation, not content. The kids know how to make things 'look nice' but they dont have anything worth saying.
Here is my take it on: remove all computers from elementary school (K -> 6th grade), add them in at the 7th grade level for basic word processing only (no powerpoint) along with a typing class. In high school add them in where the material can actually use it (physics visualizations, math, etc). Add them to the library at that level as another research tool.
Call me technophobic, but I see the use of computers in the classroom as a crutch more than as a tool that extends the students knowledge.
vservers? are you kidding? They are a joke compared to solaris zones. Network configuration is one issue. The lack of mainline linux support is the next, thus its not very well tested. Its a hack that is bolted onto the side.
Debian? thats nice, but none of the commercial apps I use support that distro. In my case, Oracle is the big one. They support RHEL, SLES, Asianux and of course, Oracle Linux. Debian is not on the list. They do support solaris zones, but not linux vserver containers.
I can more or less pick any OS I want if I have the source code to the app. But in that case I'll just compile it for solaris and be done with it.
And this brings me to the next major annoyance about 'linux'. It is not one product, it is a catch all for a dozen different, often incompatible, variants based on a common base that then go off in a dozen different directions. I have vendors supplying me with drivers that are specific to not only the distro but the patch level because things change.
If we were talking Solaris 8 or 9 vs linux, it would be no contest. Solaris 10 is another story.
In the current shop where I work, Solaris is the OS of first choice. We only run linux if we absolutely have to due to application compatibility issues. From a cost standpoint, they both run on the same server hardware (intel or amd), so there is no cost advantage there. The major one is that a RedHat support contract cost more than the equivalent one from Sun for Solaris (they are both free if you dont want support... sorta.... You can download solaris from sun for free and install it. You cant do the same with RedHat. That said, if you use kickstart you can get around the mandatory rhn registration that stops you dead if you dont have a valid support contract.)
It just that with solaris we have the ability to load the box up with dozens of zones that are easy to manage compared to the alternatives on linux (yes, there are alternatives to solaris zones, and all of them involve unsupported kernel patches. ) We tend to run 20+ zones per dual proc box. Each zone gets its own env. We dont need them to differ too much from the main zone, so things like Xen are overkill. Chroot would be nice if we could also get better control over things like IP's, admin access, hostname resolution, etc. For what we want, Zones are absolutely perfect. zfs and Dtrace just add icing to the cake.
That said, this article read like a BMW sales rep's opinion of the newest Audi. If I want opinions of if solaris is dying, I'm not going to go and ask the head of the linux consortium which has a vested interestin seeing their prophecy come true.
For a good time, go to the IDC site and read the comments. Most of them are ripping the author for the piss poor job he did since it reads like a linux marketing piece more than an actual news article.
If you were stuck with a closed OS like AIX, HPUX or Solaris you would still have the option of compiling perl from source.
The nice thing is that the _app_, not OS, is free and thus he has the option of picking another way of working around the issue.
That said, if this was a kernel issue, he would have no choice but to compile a new one in and totally void his support contract. Yes, he would have the option of doing it but then he loses the "support" that he paid for.
Uh, is this not one of the main features of the ZFS file system? It does a checksum on every block written and will reconstruct the data if an error is found? (assuming you are using either raid-z or mirroring. Otherwise it will just tell you that you had an error).
The best quote from the article was this: "I've taught a fair number of women students in electrical engineering and computer science classes over the years. I can give you a list of the ones who had the best heads on their shoulders and were the most thoughtful about planning out the rest of their lives. Their names are on files in my "medical school recommendations" directory."
We dont go through enough adds/moves/changes to make it worth the effort of tracking with anything more than a spreadsheet. Given that I've managed to cut the number of physical servers in half over the past 4 years (during which ime the number of services have gone up 10x) I dont predict the need for anything more specialized anytime soon.
I've seen some palces that turn over a a few doezn servers a day and they needed a management app to keep track of everything. I'm lucky enough to not require such complexity
I went the other route: amp meters on all of the new in rack PDU's (we use the chatsworth vertical power strips and opt for the amp meter option). Its a bit more expensive, but not too bad when you consider the overall cost of what is in that rack.
This way I know exactly how much the rack as a whole is pulling. More importantly, I know that each of the PDU's in the rack is under 50% so that in case a power supply causes a short, I dont loose the entire rack due to a cascading failure. (I've seen it happen when folks use the 'outlets are available, thus the rack can support another server' mentality and a PSU goes really bad or they just simply overload the system. Easy to do with 110V feeds)
Based on the direct power readings, our 'de-rating' is based on what we measure rather than what the PSU says. So for us, we know a typical 2 proc dual core amd based server pulls 1.3A @ 208V (about 300w). A full rack of Sun 6140 arrays's? (14 trays total) that thing pulls about 22A @ 208V.
This also helps us plan how many circuts and what size to drop into each rack. We have found that for our server density, two 30A@208v feeds does the trick. (vs 4x20A or similar). This way we save on panel space, and that makes our electricians happy (rule #1 of the datacenter, do not annoy the HVAC guy's or electricians:-)
1) We did it as part of the finish script in the kickstart script. It was more than just adding it to the grub.conf, you also have to tweak a few other files to actually get the damn thing to start a login prompt on the serial port. Compared to solaris where I just add a console=ttya in the add_install_client script (yeah, we jumpstart everything) as a boot param and the installer takes care of the rest.
2) I ended up loading the lsi mpp drivers. Using a combination of veritas dmp debug commands and the mmpUtil command, we can map which lun it comes from. the vx commands give us the uuid of the lun, and the mppUtil commands tell us what lun has that uuid. Oh yeah, and we use VxVM for the fast resync capabilities since we run a campus cluster. ZFS offers a similar function of only copying the changed blocks over when (not if) something goes wrong. I hate waiting for a 2TB volume to fully resync, and so do the DBA's who then bitch about the performance hit.
3) We first tried the magic that redhat suggested (echo "bla" into a bunch of/proc files). Then we tried the mppBusRescan utility that comes with the lsi drivers. Noe worked because they were only operating at the scsi level, not the FC port level. The missing magic? we have to first force a lip on the FC ports by echoing more stuff into more/proc files. `cfgadm -c configure` is so much easier since it does everything in one shot, and that is only if you turn off the autoconfigure option.
3) 802.1ad is bonding, aka EtherChannel in the cisco world (truning to everybody else). I mentioned that we are running between two different switches and bonding is not supported in that config. This is the problem with linux. I say I want to do something, and everybody jumps on the wrong answer. (yes, cisco will support it between 6509's running the not yet out Sup 1440's.... like I trust that to work, 10G is just easier for the bandwidth)
4) linux-vserver... its close, but zones takes it to the next level. And I hate patching the kernel since you are just asking to get something to not work since you are now one of a very small group doing it. If i wanted to do that, i'd work for redhat/suse/oracle/whoever. Yeah, it was fun in the early days of slackware when you had to more or less build a new kernel from src to do anything, but I really dont have time for that anymore.
5) ps and top are good for easy problems. Here is an one that those tools didnt spot: I had to figure out what was eating a system alive and sending it into a tailspin. Turns out that oracle's enterprise agent was spawning thousands of sub jobs that lived and died in under a second. dtrace spotted this in an instant. top and ps will never saw it. (once you know the problem, strace will also spot it, but you first have to know what process is causing the problem) Yeah, when somebody says the system is 'acting funny', I reach for top, ps, ptree and the like. If nothing shows up in 30 seconds, its time to dig deeper. That requires tools like truss, snoop, dtrace and mdb (for the really nasty problems). I dont get paid more than everybody else in my group to solve easy problems. I get paid more because I get to solve the nasty ones.
As for zfs, the first release was cool, but had a few 'issues' that needed to be sorted out. At least the sales engineers I talked to warned me to wait for an update release or two before I used it in production. That was then, but this is now, and it rocks. We now use it as the base FS for all of our zones and as a failover fs with Sun Cluster. The damn thing just works. I would not put oracle on top of it due to a few minor strange things oracle does, but for everything else it is great.
And dont feel so bad, I think poorly of HPUX and AIX too:-) Talk about two vendors who have not really bothered to advance their OS in the past 10 years, linux is awesome compared to those two. (of course, my experience with those two have been helping the poor admins try and do half of what linux can do)
Slashdot Mirror
Yes, but when I was in high school back in 1989 we had 1 classroom with PC's and a 4 of 5 in the library. So the student to computer ratio was about 1:100.
Now they want a computer in every classroom for every student. And then one on every teachers desk, and another high end lab in the library. The student to computer ratio is now approaching 2:1 in favor of the computers.
And adjusted for inflation, the cost of managing 25 standalone pc's running DOS and a copy of some math program that resides on a pair of floppies is nothing compared to the cost of 25 pc's running windows 7, connected to the internet, with few dozen apps.
Computers are great and all, but I have yet to see a good study that showed that the kids with ubiquitous computer access were better off than those who had limited computer access.
Start with an iPhone 4s or Android equiv (http://www.pcworld.com/article/241955-2/smartphone_camera_battle_iphone_4s_vs_the_android_elite.html)
Once you run into the limits of what that platform can do, then get a better camera. But by then you will have a better idea of why the phone does not work for you. If it is focus speed, go high end DSLR; if it is image quality, go 4/3rds or low end DSLR, etc.
For my chosen field (sports photography), a Nikon D300+MB10D w/ 70-200mm/2.8, 17-55/2.8 and SB-900 is entry level (sports photography) but comes at a high cost (as in $4-5k). For others, maybe a mirrorless system (4/3rds, Nikon 1, etc) would be a better option and a full system would cost less than one of my lenses.
Don't take personal offense but step back and look at the client -> agency -> contractor relationship and how fundamentally screwed up it is.
Most importantly, ask yourself if your primary purpose in being at a clients site is to make the customer happy or if it is to make your employer money.
We all want to say our primary job it is to make the customer happy, but in reality our primary job is to make the employer employer money (or else you wont be in business very long. Customers are usually happier if you don't charge them).
Now are you doing your employer (the folks who sign your checks) any favors by reducing the long term amount of work that needs to be done?
You may say counter by saying that if you do a good job then the customer will hire you back and it makes your employer more money in the long run. In reality, if the customer was qualified to determine if you did a good job (fix it permanently) vs an adequate one (patch it), then you would probably not be there to being with (otherwise they would do it themselves). As others have said in the past, "adequacy is sufficient, all else is superfluous."
Yes, this is a very Machiavellian view of the contracting world.
This has nothing to do with how you as an individual approach a given situation. You may have a higher personal ethical standard and want to go well above and beyond the call of duty. I've met a number of consultants/contractors/perms/temps/whatever who are like this. I like working with these people, mostly because they will work as a team and do what is right for the project as a whole and not just their department or group.
But I also have to face the reality that the contracting environment as it is setup today is fundamentally flawed in that it financially encourages mediocre work.
On average I'd say 2/3rd fail to make it past the first month. We have had some that didn't even make it past the first week. (HINT: do not make sexist comments to your female coworker. )
We don't try and trap them, but it seems that their inflated sense of self worth and job skills are such that we think that they actually go out of their way to screw up and then argue with us about how they didn't screw up. We typically give them a 2nd chance to fix it once we point it out to them. The thought is that if they can learn from their mistakes then they have hope. If not, well.... Next...
I see reading comprehension is not one of your strong points, but I'm happy to say that you are very good at jumping to conclusions.
As I said before, as a fed contractor, I was restricted to working more than that. I could want to have be there for 80, but the higher ups told me to not screw up the accounting, and not to even think about going off clock since an audit would nail them. When I left they back fill me with two new folks. My former co-workers bitched about my leaving, but the 50% pay raise was too much to ignore and I was bored out of my mind there.
So I jumped to a consulting firm in the middle of the dot com boom days. In one year I billed almost 3000 hours. Take out the vacation and 2 weeks from having a kid and I'm over 70 as an average. Add in the crap that was off the clock and it is higher.
If you have been around for 20+ years you should know that on average you get most of your work done in the first 20 hours of the work week. After that its an exponential decay. After 60 you are just taking up space and consuming coffee. (The occasional crunch time is one thing, but I'm talking averages). But when you are a consultant billing $300/hr, the company who is cutting your checks really does not care as long as the customer is happy.
So much for your assessment of my abilities.
And the odd thing is that the group that I now work with in a local county government (all direct) is by far the best damn group I have ever seen or had to work with. Based on your off the cuff assessment, we should not exist. We should all want to leave for more money and the thrill of working long hours.
What you fail to understand is that it not the type of job you are in. I've seen good and bad groups of contractors, consultants, government workers, education, commercial, etc. I've run into incredible people in places I least expected to. I've also run into some real idiots in spots where you wonder how they ever survived there. You name an industry or group and I'm sure somebody will toss out an example of a good team and an example of a crap team. In the end it is the people you work for and with. Good people want to work with other good people. Its a positive feedback cycle. A good group knows when they are talking to other good people. The flip side is also true. Bad people work with other bad people (and the dunning-kruger effect is in full effect in those environments)
And please don't confuse productivity and quality with 60+ hour work weeks, or even 45+. We had a security guy who did 70+ every week. He was a hard worker and knew the talk but fundamentally lacked any real skills. When he left to become a consultant we replaced him with a few small shell scripts. (he spent 20+ hours just creating new user accounts; by hand in ldap. One guy wrote a script and it now takes us 10 minutes a week, if that) The firm he now works for has no clue that he has no clue, because the entire company has no clue. But the owner knows some folks with three letter titles from the golf course and they get security assessment gigs because of it. C'est la vie.
Disclamer: I was a salaried contractor who worked for Northrop who was sub for Lockhead who worked for the USAF. A typical oversized, multi-year, multi-billion dollar contract.
Northrop could not replace me at any time they wanted without cause any easier than my current government employer could. (my current employer is a local county.) Nobody in that entire building ever got fired "just because." Its often bantered about, but given that most of the contractors were ex-military in their 40's (I was one of two exceptions) it never happened. Once they got you in there, it was simply too expensive to swap you unless you royally screwed up and the customer wanted you out of there. And I got full benefits from Northrop, including vacation, sick, medical, 401k, etc, etc. I get more or less the exact same benefits from my current employer (401k, no pension).
I was salaried, but I had to declare all of my hours to the project for billing purposes. They did not want me to burn hours faster than what was allotted in the contract, so they really didn't want me there more than 40 unless something major came up. My current job is also salaried, and on average I probably work the same average number of hours.
Now, individual situations can alter this perception. Maybe you worked for SAIC or some other warm body shop in a contract from hell that had high turnover. But in all of those cases, the issue was not that folks got fired, it's that they all left the second they found a better paying job because the contract went to the lowest bidder, and SAIC only wanted to pay 1/3rd to 1/4 of the rate to the actual person doing the job to cover their "overhead" ($60/hr for a system admin to SAIC results in $20/hr to the actual contractor. Any takers for a unix admin with a security clearance for @$30-40k/yr? No? Didn't think so.).
We do use contractors/consultants at my current job in two cases:
1) Very short term 'in, do xyz, get out' implementation or staff augmentation projects. Basically, install this software, train the IT staff and get out. Those are all $250+/hr consultants and we limit how long they are in house. They are very expensive ($500k/yr) but we only use them to get over critical staff shortages or crunch periods.
2) Contract to hire for new folks at the help desk. Mostly because we can, er, fire them faster. That said we have yet to do it "just because." In all cases the idiot screwed up not once, not twice but three times within the first two weeks and got walked. But when you get here those conditions are very explicit. Going through the interview phase is a pain in the ass for us (we'd rather be doing real work) so once somebody is here we'd rather it work out than not. It actually cost us more to go that route but the process is faster so nobody cares.
Having worked in education, dot.com startups (during the boom and bust years), federal (DoD) and now Local government, I have found that the desire to "do things properly'"is not a trait of the type of job (contract, perm, etc), it is a trait of the individual.
Sorry, I've worked both sides of this fence, and you should have stopped before you even typed the first word.
The contractors for the fed/military/etc do not work day to day, twice the hours or have triple the productivity.
They are given year+ long contracts, work the same hours, and have the same or less productivity. The perms face the exact same thing, their entire division can be wiped away with the stroke of a budgeting pen.
We are not talking about day labors here, all federal contracts are long and well defined. While your project may get canceled with the next _YEARLY_ budget, the odds of it suddenly going under are next to 0.
As a fed contractor, I never put in more than 40 hours a week. That is what we had in the budget, and to do more than that would have resulted in issues. The "cost+" contracts that would let me work 80 hour weeks and have the contracting agency get paid for it are few and far in between. Most are fixed at the rates and the number of hours, it does them no good to have you work more than your scheduled rate.
The productivity thing is pure bullshit. I've seen incompetent admins on both sides, but most are on the contracting side because the contracting firm wants to keep a larger % of the cut to themselves, and thus toss inexperienced newbies into the slot in the hopes that nobody will notice. The real kicker is that as a contractor you have an incentive to not really fix things, but to just patch them. After all, why fix something once and for all when your job depends on the customer needing to have you around to constantly fix something?
To which Sabine Schmitz stated "I could do that in a transit van..."
http://www.youtube.com/watch?v=BQJKQjXpGQA
(she almost did it...)
But its not just by zip: its zip, street and address. Tax districts can vary house by house. 100 Main St. can be in a different tax district than 101 Main St. We have a road named, aptly enough, 'County Line Road'. Homes on one side are in one county, homes on the other side a different county. Both use the same zip code because the post office does not care.
Now as a business you can use a simplified zip = X% mapping but if you under collect then you are screwed. (If you over collect nobody cares).
Back in the day companies could get away a simple lookup table and hope for the best simply because nobody was really looking. We (fyi: I work for a county) now have an app that integrates with our gis parcel data and geocodes the entire thing and then sends the State a bill for what they owe us based on what they collected. In its first month of use the app has 'saved' us about a 100k in what was unclaimed funds because in the past we were just keying off of the zip code.
There is, but it involves geocoding every single address. And then updating it every time any one of the 60,000 tax districts change their boundaries or rates.
Here is the problem, you can have two houses on opposite sides of the street be in two different tax districts. So a simple 'if zip == xxxxx, then tax = Y' type of lookup table will not work.
You then have the issue of the corporation needing to potentially apply for a sales tax license in jurisdiction before they can collect the tax.
Then you have the issue of having to possibly send the check to 3 or 4 different groups on different schedules for each customer in a different.
And finally there is the question of what gets taxed. In some states, some items are not taxed (usually basic food). So if I order a 10lb tub of powdered gatorade from amazon.com it may get taxed in one state but not another, both of which have a sales tax.
To call it a mess is an understatement. This is the main reason why the courts tossed out the states requirement to collect the tax: the burden was simply too much. If memory serves me correctly, that same court decision left the door open to enact a simplified sales tax scheme (if shipping to NY, then charge X% and send it to Y address and be done with it).
See Jonathan Lethem's excellent essay on the subject: http://harpers.org/archive/2007/02/0081387
Bonus points if you get to the end and read through the references section :-)
Same here.
I picked up an HP LaserJet 4100 (with duplexer and jetDirect card) for $25 at a used pc sale run by the local county government. I also got an HP 8150DN (duplexer, network and 2000 sheet feeder tray) at the same sale for another $25.
The things are built well and everybody supports them. Because they were so common, toner is easy to find. Not that I'll need it, the 8150 came with two full cartridges rated at 20,000 sheets each.
Oh, and nobody goes on holiday without contact for over 24 hours, do they? I bring a laptop and a smartphone with me wherever I go. Even when I visited Northern Africa, I made sure to get online at least once a day to check, act on, and reply to my email.
Its not a vacation if you can find me.
I leave my cell, laptop, etc home. For my last trip, I told my co-workers what park I would be in and that if something went south that they can call the park ranger and then hope that they can find me.
I want to get away from the the regular grind, not bring them with me :-)
Technically, at this time oracle does not own sun.
They have announced that they will purchase them and the sale is pending, but until that time the two companies are totally independent and functionally must continue to operate as such.
So sometime this summer the oracle logo will be correct, but currently it is wrong.
Cisco + Sun would make more sense. Mostly because there is very little overlap in their actual products but their two lines constantly need to work together. (Our sun servers are connected to Cisco ethernet switches, our SunRays vpn into Cisco vpn concentrators, our Sun Storage is connected to Cisco MDS switches, etc). It would also give Cisco the biggest, baddest InfiniBand switch on the market (and at 110Tbps, its switching capacity totally trashes anything cisco has ever produced).
The biggest problem with the Sun+IBM deal was that there was so much overlap, customers would be left to wonder which product lines would get discontinued. (glassfish vs websphere, solaris vs aix, sparc vs power, sun's servers vs ibm's, storage, tape, etc, etc, etc. )
I'm thinking along the same lines.
I never touched a computer till the 7th grade, and never did anything more than basic word processing with them till the 10th grade. I learned how to do calculus using chaulkboards and paper.
My kid is in the 2nd grade and already doing powerpoint. WTF?!?!?! The focus is on presentation, not content. The kids know how to make things 'look nice' but they dont have anything worth saying.
Here is my take it on: remove all computers from elementary school (K -> 6th grade), add them in at the 7th grade level for basic word processing only (no powerpoint) along with a typing class. In high school add them in where the material can actually use it (physics visualizations, math, etc). Add them to the library at that level as another research tool.
Call me technophobic, but I see the use of computers in the classroom as a crutch more than as a tool that extends the students knowledge.
vservers? are you kidding? They are a joke compared to solaris zones. Network configuration is one issue. The lack of mainline linux support is the next, thus its not very well tested. Its a hack that is bolted onto the side.
Debian? thats nice, but none of the commercial apps I use support that distro. In my case, Oracle is the big one. They support RHEL, SLES, Asianux and of course, Oracle Linux. Debian is not on the list. They do support solaris zones, but not linux vserver containers.
I can more or less pick any OS I want if I have the source code to the app. But in that case I'll just compile it for solaris and be done with it.
And this brings me to the next major annoyance about 'linux'. It is not one product, it is a catch all for a dozen different, often incompatible, variants based on a common base that then go off in a dozen different directions. I have vendors supplying me with drivers that are specific to not only the distro but the patch level because things change.
I agree.
If we were talking Solaris 8 or 9 vs linux, it would be no contest. Solaris 10 is another story.
In the current shop where I work, Solaris is the OS of first choice. We only run linux if we absolutely have to due to application compatibility issues. From a cost standpoint, they both run on the same server hardware (intel or amd), so there is no cost advantage there. The major one is that a RedHat support contract cost more than the equivalent one from Sun for Solaris (they are both free if you dont want support... sorta.... You can download solaris from sun for free and install it. You cant do the same with RedHat. That said, if you use kickstart you can get around the mandatory rhn registration that stops you dead if you dont have a valid support contract.)
It just that with solaris we have the ability to load the box up with dozens of zones that are easy to manage compared to the alternatives on linux (yes, there are alternatives to solaris zones, and all of them involve unsupported kernel patches. ) We tend to run 20+ zones per dual proc box. Each zone gets its own env. We dont need them to differ too much from the main zone, so things like Xen are overkill. Chroot would be nice if we could also get better control over things like IP's, admin access, hostname resolution, etc. For what we want, Zones are absolutely perfect. zfs and Dtrace just add icing to the cake.
That said, this article read like a BMW sales rep's opinion of the newest Audi. If I want opinions of if solaris is dying, I'm not going to go and ask the head of the linux consortium which has a vested interestin seeing their prophecy come true.
For a good time, go to the IDC site and read the comments. Most of them are ripping the author for the piss poor job he did since it reads like a linux marketing piece more than an actual news article.
If you were stuck with a closed OS like AIX, HPUX or Solaris you would still have the option of compiling perl from source.
The nice thing is that the _app_, not OS, is free and thus he has the option of picking another way of working around the issue.
That said, if this was a kernel issue, he would have no choice but to compile a new one in and totally void his support contract. Yes, he would have the option of doing it but then he loses the "support" that he paid for.
Uh, is this not one of the main features of the ZFS file system? It does a checksum on every block written and will reconstruct the data if an error is found? (assuming you are using either raid-z or mirroring. Otherwise it will just tell you that you had an error).
The best article written about this was by Philip Greenspun (MIT Prof) at http://philip.greenspun.com/careers/women-in-science
The best quote from the article was this: "I've taught a fair number of women students in electrical engineering and computer science classes over the years. I can give you a list of the ones who had the best heads on their shoulders and were the most thoughtful about planning out the rest of their lives. Their names are on files in my "medical school recommendations" directory."
Keep track? hahahahah Its all up in my head :-)
We dont go through enough adds/moves/changes to make it worth the effort of tracking with anything more than a spreadsheet. Given that I've managed to cut the number of physical servers in half over the past 4 years (during which ime the number of services have gone up 10x) I dont predict the need for anything more specialized anytime soon.
I've seen some palces that turn over a a few doezn servers a day and they needed a management app to keep track of everything. I'm lucky enough to not require such complexity
I went the other route: amp meters on all of the new in rack PDU's (we use the chatsworth vertical power strips and opt for the amp meter option). Its a bit more expensive, but not too bad when you consider the overall cost of what is in that rack.
:-)
This way I know exactly how much the rack as a whole is pulling. More importantly, I know that each of the PDU's in the rack is under 50% so that in case a power supply causes a short, I dont loose the entire rack due to a cascading failure. (I've seen it happen when folks use the 'outlets are available, thus the rack can support another server' mentality and a PSU goes really bad or they just simply overload the system. Easy to do with 110V feeds)
Based on the direct power readings, our 'de-rating' is based on what we measure rather than what the PSU says. So for us, we know a typical 2 proc dual core amd based server pulls 1.3A @ 208V (about 300w). A full rack of Sun 6140 arrays's? (14 trays total) that thing pulls about 22A @ 208V.
This also helps us plan how many circuts and what size to drop into each rack. We have found that for our server density, two 30A@208v feeds does the trick. (vs 4x20A or similar). This way we save on panel space, and that makes our electricians happy (rule #1 of the datacenter, do not annoy the HVAC guy's or electricians
Here goes:
/proc files). Then we tried the mppBusRescan utility that comes with the lsi drivers. Noe worked because they were only operating at the scsi level, not the FC port level. The missing magic? we have to first force a lip on the FC ports by echoing more stuff into more /proc files. `cfgadm -c configure` is so much easier since it does everything in one shot, and that is only if you turn off the autoconfigure option.
:-) Talk about two vendors who have not really bothered to advance their OS in the past 10 years, linux is awesome compared to those two. (of course, my experience with those two have been helping the poor admins try and do half of what linux can do)
1) We did it as part of the finish script in the kickstart script. It was more than just adding it to the grub.conf, you also have to tweak a few other files to actually get the damn thing to start a login prompt on the serial port. Compared to solaris where I just add a console=ttya in the add_install_client script (yeah, we jumpstart everything) as a boot param and the installer takes care of the rest.
2) I ended up loading the lsi mpp drivers. Using a combination of veritas dmp debug commands and the mmpUtil command, we can map which lun it comes from. the vx commands give us the uuid of the lun, and the mppUtil commands tell us what lun has that uuid. Oh yeah, and we use VxVM for the fast resync capabilities since we run a campus cluster. ZFS offers a similar function of only copying the changed blocks over when (not if) something goes wrong. I hate waiting for a 2TB volume to fully resync, and so do the DBA's who then bitch about the performance hit.
3) We first tried the magic that redhat suggested (echo "bla" into a bunch of
3) 802.1ad is bonding, aka EtherChannel in the cisco world (truning to everybody else). I mentioned that we are running between two different switches and bonding is not supported in that config. This is the problem with linux. I say I want to do something, and everybody jumps on the wrong answer. (yes, cisco will support it between 6509's running the not yet out Sup 1440's.... like I trust that to work, 10G is just easier for the bandwidth)
4) linux-vserver... its close, but zones takes it to the next level. And I hate patching the kernel since you are just asking to get something to not work since you are now one of a very small group doing it. If i wanted to do that, i'd work for redhat/suse/oracle/whoever. Yeah, it was fun in the early days of slackware when you had to more or less build a new kernel from src to do anything, but I really dont have time for that anymore.
5) ps and top are good for easy problems. Here is an one that those tools didnt spot: I had to figure out what was eating a system alive and sending it into a tailspin. Turns out that oracle's enterprise agent was spawning thousands of sub jobs that lived and died in under a second. dtrace spotted this in an instant. top and ps will never saw it. (once you know the problem, strace will also spot it, but you first have to know what process is causing the problem) Yeah, when somebody says the system is 'acting funny', I reach for top, ps, ptree and the like. If nothing shows up in 30 seconds, its time to dig deeper. That requires tools like truss, snoop, dtrace and mdb (for the really nasty problems). I dont get paid more than everybody else in my group to solve easy problems. I get paid more because I get to solve the nasty ones.
As for zfs, the first release was cool, but had a few 'issues' that needed to be sorted out. At least the sales engineers I talked to warned me to wait for an update release or two before I used it in production. That was then, but this is now, and it rocks. We now use it as the base FS for all of our zones and as a failover fs with Sun Cluster. The damn thing just works. I would not put oracle on top of it due to a few minor strange things oracle does, but for everything else it is great.
And dont feel so bad, I think poorly of HPUX and AIX too