Bullying sales, whilst effective, is fundamentally a workaround for the underlying problem. Not to say it doesn't work, because as you say, they have power - sales brings in money. Big customer gives money. Big customer expresses disastisfaction, and then it's worth real money to sort it.
I just feel a little dirty, each time I do it.
Come to think of it, that might be more to do with dealing with salespeople, than what I actually talk to them about.
Cultural differences can also make things like tech support difficult. The 'way things are' in an Arabic country are just different. No, that's _not_ worse, just different. However when you're trying to interface between any two things, then... well everything works better if it's as simple as possible.
This goes double in a troubleshooting situation. It doesn't matter much if the 'interface' is your TCP/IP connection, the right ODBC drivers, or just being on the same damn timezone and language as the person you're dealing with.
I'm aware of a corporate call centre that's move to hungary. Eastern Europe is proving increasingly popular. Same problem I'm afraid. I'm entirely prepared to accept and appreciate that the guy on the other end of the line knows his stuff. At least, until he's proved otherwise. However I get very frustrated when I can't follow English + accent + bad line + 'did you reboot it' tech support.
I've experienced both good and bad support. I know how good the 'good' can be. And likewise how bad it can be.
The problem is, a) price - 'end users' want free. Business customers want cheap. The latter often can be persuaded with an argument of 'business impact' of crap support.
b)End users lie. I mean, seriously. They're _terrible_ at admitting that they did something. I've seriously had someone bring a laptop to our helpdesk, with tyre tread marks over it. And when asked 'so, what happened to that then' his response was the usual defensive 'nothing, it just stopped working'.
c) Support work sucks. I mean, seriously, I would have to be desparate to work in a call centre. It's a crap job on the coalface. Even if the payscales were good, I'd be very wary of it. (and by good, I'm taking 'suck it up for a couple of years, then go buy a house' level of good. I have had that discussion, there's some high power high profile jobs that pretty much relies on a couple of years running near burnout). So... you get the people who are there, either because their desparate (and will leave as soon as they can) or crap.
d)Home systems really lack badly in diagnostics. Servers, especially high end Unix, it's entirely possible to get extensive diagnostic information. I mean, you can plug into the diag processor of a Sun, and have it tell you precisely why it's not powering up. If it's 'up' you can do all manner of kernel hooking, especially with ultra-sexy tools like dtrace, which mean you can actually find out what's wrong. There is a reason why you go for the '3Rs' with windows. (Reboot, reinstall, rebuild)
Call centres _can_ be 'reasonable' but... doing so gets very expensive. That cost is compared in direct competition with el-cheapo helldesk in the middle of nowhere, and you lose out, since they don't actually put some considerationg into ever actually having to _use_ the support in a critical emergency situation.
I find my mobile invaluable. It's really great for playing the 'herding cats' game, with a bunch of friends. Ring up a bunch of people, see if they feel like going to the pub, that kind of thing. SMS especially, is far less intensive than 10 separate phone calls.
I also find it useful for impromptu meetings - phone someone up, because you're in town and feel like saying hi, to see if they're available.
I can make do without PDA, laptop, and the 'extra bells and whistles' on my phone, but mobile comms I find phenomenally useful.
Man hours 'wasted' from junk emails, also adds up to quite a disgusting amount each year
Having a piece of litter dropped in your garden very occasionally, is irritating, but takes very little effort to sort out. However when it's 30 pieces of litter a day, then it becomes irritating and timewasting, and a major nuisance. I seem to recall a quote of some 90 billion spams sent, and a third of the total bandwidth of the 'net. That's really not a trivial matter.
But it wasn't one spam. That was the number of users in AOL who got spammed.
I'd instead ask: How many people _haven't_ had 5 minutes of their life wasted by spam?
That is, of course, assuming that it was only ever AOL users who got spammed by this fine gentleman. And that the only 'cost' in email is the time taken to read it, decide it was spam, and delete it. Which isn't the case. Bandwidth costs, as does high performance disk. The volumes of spam are such that quite a few places are having to upgrade hardware and add additional servers to deal with that workload
Lets not be forgetting the deeply sick proportion of the bandwidth of the world that's taken up by spam. ISTR it was somewhere around the 1/3rd of all internet traffic mark. That's one hell of a lot of bandwidth wasted. Bandwidth that's not cheap at all, especially when you start talking about transatlantic communications.
Or perhaps the collective time of the people involved to filter out the incoming junk. I see at least 1000 per month caught by my filter. A filter that _used_ to be entirely unnecessary.
Having an active email account on the internet, almost guarantees getting spammed. OK, so I can tidy it, delete it or otherwise remove it. Much like I can pick up the empty beer cans that someone has decided to drop in my garden. This doesn't mean I appreciate it in the slightest.
This spamking has made a very larger sum of money indeed, by some seriously antisocial behaviour.
No, because you're so busy concentrating on the punching and kicking, you forget about the falling part, and end up flying. An admirable way to throw yourself at the ground and miss.
From experience EMC kit is very expensive, but it is worth it. In a high end business environment, the level of tolerable downtime is zero. I mean, I'm working supporting a big name UK bank. Our 'support obligations' involve a 45 minute 'time to fix' on a critical incident. Yes, that's including the time it takes me to get out of bed and have my coffee.
Our contract was negotiated with that in mind. It was Really Expensive, because... well in IT anything's possible, but the price tag can get extreme. So we run EMC storage arrays, with very expensive support. It's not that I _couldn't_ build the same amount of storage out of JBODs or other SAN configuration, it's quite simply that we _need_ to be able to pull things back together very quickly, or face very significant penalties.
We run multiple arrays, replicating between them, to remote sites, and agressively replace any components that 'might fail soon'. It's expensive, and intensive, but compared to the cost of 'outage' (or worse, data loss) it's nothing. When you stand to be fined multiple millions for 'problems' by the regulators, it's in your interest to invest heavily in your toys.
Personally, I couldn't afford EMC's support (or any of the other vendors, but I've only really used EMC extensively) but having dealt with them on a lot of occasions, they're one of the few that are actively useful to be talking to.
And for my money, ZFS isn't really a replacement for SAN or NAS. They do similar things, true, but the overlap isn't total. *shrug* I'd say pick out the right tool for the job, but keep in mind that these things can be used in a complimentary fashion.
Full mirroring is good for that huggy feeling of data resilience, but it really does start to add up very quickly. It's not just the price of the drive that's the issue, it's the number of array enclosures, space and power that start to add up as well.
Personally, for most solutions, I find RAID5 entirely acceptable. Lower overhead, but still with resilience.
I have more difficulty with the 'mind state' transportation.
I mean, your brain doesn't ever shutdown and quiesce it's activity, so you're always going to have 'information in transit' down your synapses.
So, not only do you have to be able to measure (heisenberg willing) the state of every atom in the body, you also have to do so in an infinitesimal amount of time, or risk data loss.
Even if we get as far as actually being able to transfer 'matter replication' then there's still the fairly fundamental problem of the 'state' of the thing you're transporting.
I mean, it's hard enough getting 'consistent' backups of computer systems (which basically requires suspending operations to get a good copy). We still have no clue how to define the 'state' of a human system. It's not like we ever have to shutdown, and restart people.
non attending interviews is a dangerous game. I mean, the recruiter gets pegged, but so do you. *shrug* if you're genuinely sure you'll never want to work for that company, ever, in any role at all, then fine, go for it.
Because invariably you have a different person writing it in the first place, to the person who has to maintain it. A contractor that's intent on filling the minimum spec he can get away with, at a horrific dayrate, before 'handing it over' to go wrong and be horrific.
Also, given beauty is in the eye of the beholder, no one ever sees their code as 'disgusting'.
Having seen the difference between 'challenging' and 'sane' working hours, I have to say, there's a _reason_ why most places do a 37.5-40 hour working week. It's because quite simply, that despite time and a half, and on-call pay, there's much more to life than that. I'm on call at the moment. It's an extra... well useful chunk in my pay packet. £200 or so, + time x1.5 for any calls. Given the option of not doing it though, I'd rather just stick on a 'normal' working day when I can go home at quittin' time, and forget about work.
This seems to be the case in IT much more than other career areas. OK, so you might get phoned if your office burns down or something, but the regular requirement to be near home, with your laptop and mobile switched on, I think is mostly a delight of the IT industry.
I'm curious. You're in a minority doing what you do. But... well, is there a particular reason you can discern? I mean, is it due to discrimination (be it cultural or not)? I always assumed (and yes, I know assuming's bad) is that different genders just have different mindsets, and different things they 'tend to like'. I've ran into a minority of women working in IT. There's a few who seem to have been hired for the 'novelty value', but... well, in general I don't see much indication of differences of ability. Admittedly that's with a fairly small sample group.
What about it guys? Do you have little problem sacrificing your integrity to work in IT?
No, not really.
Let me set that in context. I work in a IT for finance environment, that's got all manner of regulation and control, and constraints. I know there's a lot of stuff that we're 'just doing wrong'. In fact, I'd go as far as saying we've a hell of a lot of stuff that's really just not very good.
I also have a fairly good idea of how to fix some of it. Unfortunately, rather a lot of these go as far as 'throw it away, and start again'. As I'm sure you're aware, that kind of approach is often... well limited for all sorts of reasons.
So I content myself with scribbling down a solution to a problem. Then deciding whether it's something I'll be able to implement or not. Most fall into the category of 'no one's going to pay because you think the UI on app XYZ is moronic' levels of problems - staff here are well trained into thinking that their IT is supposed to be complicated, obfuscate and just plain stupid to use.
So there's no real perceived need to change, and I have neither time, nor a huge amount of enthusiasm to do a 'cost justification' of exactly why using a _real_ database rather than an email system to pretend to be a database is a good idea.
So I lay aside my integrity, my knowledge of how things could and should be, in favour of the pragmatic. I have a job to do. I don't have the power to fix these problems in the 'right' way, I just have to workaround them as best I can.
I use this service, and it's great. However I've noticed there's quite a few situations where it doesn't work all that well. Simply because it just doesn't really cover 'urban' or 'not exactly motorway' roads.
I think a combination of analysing movement patterns on particular routes at particular times and days, combined with real time statistics on how fast traffic is moving would be really useful. Yes, some will get caught in congestion still, but...
I'm fairly sure though, that my TomTom, whilst it downloads traffic updates, doesn't do it based on 'reporting' of peers.
Now, if you could set it up so the 'peers' communicated with aggregated flow information, then that would really be a -very- useful trick. 'cars slowing in 2 miles, average velocity 10mph'. 'numerous vehicles stationary between J5 and J6 on road XYZ, congestion or accident or something'.
Or even just as simple as 'road's getting busy, and slowing down, might want to go a different way today'.
C remains popular, simply because there's a direct correlation between machine code instruction sets, and C. I think it'll stick around for precisely that reason.
I mean, your programs end up as machine code right at the bottom level. The chain of machine code -> assembly -> C -> higher level language I think remains as valid as it always did.
Slashdot Mirror
I just feel a little dirty, each time I do it.
Come to think of it, that might be more to do with dealing with salespeople, than what I actually talk to them about.
This goes double in a troubleshooting situation. It doesn't matter much if the 'interface' is your TCP/IP connection, the right ODBC drivers, or just being on the same damn timezone and language as the person you're dealing with.
I've experienced both good and bad support. I know how good the 'good' can be. And likewise how bad it can be.
The problem is, a) price - 'end users' want free. Business customers want cheap. The latter often can be persuaded with an argument of 'business impact' of crap support.
b)End users lie. I mean, seriously. They're _terrible_ at admitting that they did something. I've seriously had someone bring a laptop to our helpdesk, with tyre tread marks over it. And when asked 'so, what happened to that then' his response was the usual defensive 'nothing, it just stopped working'.
c) Support work sucks. I mean, seriously, I would have to be desparate to work in a call centre. It's a crap job on the coalface. Even if the payscales were good, I'd be very wary of it. (and by good, I'm taking 'suck it up for a couple of years, then go buy a house' level of good. I have had that discussion, there's some high power high profile jobs that pretty much relies on a couple of years running near burnout). So ... you get the people who are there, either because their desparate (and will leave as soon as they can) or crap.
d)Home systems really lack badly in diagnostics. Servers, especially high end Unix, it's entirely possible to get extensive diagnostic information. I mean, you can plug into the diag processor of a Sun, and have it tell you precisely why it's not powering up. If it's 'up' you can do all manner of kernel hooking, especially with ultra-sexy tools like dtrace, which mean you can actually find out what's wrong. There is a reason why you go for the '3Rs' with windows. (Reboot, reinstall, rebuild)
Call centres _can_ be 'reasonable' but ... doing so gets very expensive. That cost is compared in direct competition with el-cheapo helldesk in the middle of nowhere, and you lose out, since they don't actually put some considerationg into ever actually having to _use_ the support in a critical emergency situation.
I have a phone and a laptop. They're both needed for work. Mobile phone though, I find generally useful to be able to co-ordinate social activities.
I also find it useful for impromptu meetings - phone someone up, because you're in town and feel like saying hi, to see if they're available.
I can make do without PDA, laptop, and the 'extra bells and whistles' on my phone, but mobile comms I find phenomenally useful.
Does he get a replacement keyboard? I mean, I figure with that kind of quantity, you'd have some serious issues with your 'delete' key wearing out ...
High performance disk is expensive
Man hours 'wasted' from junk emails, also adds up to quite a disgusting amount each year
Having a piece of litter dropped in your garden very occasionally, is irritating, but takes very little effort to sort out. However when it's 30 pieces of litter a day, then it becomes irritating and timewasting, and a major nuisance. I seem to recall a quote of some 90 billion spams sent, and a third of the total bandwidth of the 'net. That's really not a trivial matter.
I'd instead ask: How many people _haven't_ had 5 minutes of their life wasted by spam?
That is, of course, assuming that it was only ever AOL users who got spammed by this fine gentleman. And that the only 'cost' in email is the time taken to read it, decide it was spam, and delete it. Which isn't the case. Bandwidth costs, as does high performance disk. The volumes of spam are such that quite a few places are having to upgrade hardware and add additional servers to deal with that workload
Or perhaps the collective time of the people involved to filter out the incoming junk. I see at least 1000 per month caught by my filter. A filter that _used_ to be entirely unnecessary.
Having an active email account on the internet, almost guarantees getting spammed. OK, so I can tidy it, delete it or otherwise remove it. Much like I can pick up the empty beer cans that someone has decided to drop in my garden. This doesn't mean I appreciate it in the slightest.
This spamking has made a very larger sum of money indeed, by some seriously antisocial behaviour.
No, because you're so busy concentrating on the punching and kicking, you forget about the falling part, and end up flying. An admirable way to throw yourself at the ground and miss.
Our contract was negotiated with that in mind. It was Really Expensive, because ... well in IT anything's possible, but the price tag can get extreme. So we run EMC storage arrays, with very expensive support. It's not that I _couldn't_ build the same amount of storage out of JBODs or other SAN configuration, it's quite simply that we _need_ to be able to pull things back together very quickly, or face very significant penalties.
We run multiple arrays, replicating between them, to remote sites, and agressively replace any components that 'might fail soon'. It's expensive, and intensive, but compared to the cost of 'outage' (or worse, data loss) it's nothing. When you stand to be fined multiple millions for 'problems' by the regulators, it's in your interest to invest heavily in your toys.
Personally, I couldn't afford EMC's support (or any of the other vendors, but I've only really used EMC extensively) but having dealt with them on a lot of occasions, they're one of the few that are actively useful to be talking to.
And for my money, ZFS isn't really a replacement for SAN or NAS. They do similar things, true, but the overlap isn't total. *shrug* I'd say pick out the right tool for the job, but keep in mind that these things can be used in a complimentary fashion.
Personally, for most solutions, I find RAID5 entirely acceptable. Lower overhead, but still with resilience.
So, not only do you have to be able to measure (heisenberg willing) the state of every atom in the body, you also have to do so in an infinitesimal amount of time, or risk data loss.
I mean, it's hard enough getting 'consistent' backups of computer systems (which basically requires suspending operations to get a good copy). We still have no clue how to define the 'state' of a human system. It's not like we ever have to shutdown, and restart people.
It was a nagging itch for a while, that reference in Enders game to 'found the name in a book somewhere'. Now I know which book it was from.
Not if it's a quantum state receiver thingummy that hasn't been invented yet.
non attending interviews is a dangerous game. I mean, the recruiter gets pegged, but so do you. *shrug* if you're genuinely sure you'll never want to work for that company, ever, in any role at all, then fine, go for it.
In IT, they call that 'MS Exchange'.
Also, given beauty is in the eye of the beholder, no one ever sees their code as 'disgusting'.
This seems to be the case in IT much more than other career areas. OK, so you might get phoned if your office burns down or something, but the regular requirement to be near home, with your laptop and mobile switched on, I think is mostly a delight of the IT industry.
I'm curious. You're in a minority doing what you do. But... well, is there a particular reason you can discern? I mean, is it due to discrimination (be it cultural or not)? I always assumed (and yes, I know assuming's bad) is that different genders just have different mindsets, and different things they 'tend to like'. I've ran into a minority of women working in IT. There's a few who seem to have been hired for the 'novelty value', but ... well, in general I don't see much indication of differences of ability. Admittedly that's with a fairly small sample group.
Let me set that in context. I work in a IT for finance environment, that's got all manner of regulation and control, and constraints. I know there's a lot of stuff that we're 'just doing wrong'. In fact, I'd go as far as saying we've a hell of a lot of stuff that's really just not very good.
I also have a fairly good idea of how to fix some of it. Unfortunately, rather a lot of these go as far as 'throw it away, and start again'. As I'm sure you're aware, that kind of approach is often ... well limited for all sorts of reasons.
So I content myself with scribbling down a solution to a problem. Then deciding whether it's something I'll be able to implement or not. Most fall into the category of 'no one's going to pay because you think the UI on app XYZ is moronic' levels of problems - staff here are well trained into thinking that their IT is supposed to be complicated, obfuscate and just plain stupid to use.
So there's no real perceived need to change, and I have neither time, nor a huge amount of enthusiasm to do a 'cost justification' of exactly why using a _real_ database rather than an email system to pretend to be a database is a good idea.
So I lay aside my integrity, my knowledge of how things could and should be, in favour of the pragmatic. I have a job to do. I don't have the power to fix these problems in the 'right' way, I just have to workaround them as best I can.
I think a combination of analysing movement patterns on particular routes at particular times and days, combined with real time statistics on how fast traffic is moving would be really useful. Yes, some will get caught in congestion still, but ...
Now, if you could set it up so the 'peers' communicated with aggregated flow information, then that would really be a -very- useful trick. 'cars slowing in 2 miles, average velocity 10mph'. 'numerous vehicles stationary between J5 and J6 on road XYZ, congestion or accident or something'.
Or even just as simple as 'road's getting busy, and slowing down, might want to go a different way today'.
I mean, your programs end up as machine code right at the bottom level. The chain of machine code -> assembly -> C -> higher level language I think remains as valid as it always did.