I second that. On of the things we ask for with candidates is that they submit some code examples before we decide on interviewing them. That usually weeds people out quite well.
As well as the usual questions which only skim the surface and don't really tell you much about ability only about what they can remember. We give applicants some fairly open ended programming excersises and see what they do.
things like factoring some prime numbers and working with roman numerals.
then we have a 3 month period in which to evaluate them further. I like to set them some tasks to see what they are made of and how adaptable they are as well as how well they communicate in a team environment. I usually give them a 'prototype' style application and get them to extend it and make it more extensible for future versions and keeping the same coding style. There are several ways to solve the problem and each have there merits and weaknesses which give something to talk about the task.
The open university do a few good ones here in the uk (like the Atom series) and the history of maths shown on the bbc.
They don't go into 'serious' depth and the history of maths didn't actually cover much of the maths details but it did cover the the theroies and the history quite well.
You may be able to find them on a torrent site. (and if your really lucky you may be able to find some of those late night open university broadcasts).
There not really strictly tech but they do cover the science tech is based on.
go lookup linq etc.., there's a lot that Microsoft are doing in.net to make it a lot more parallel.
also you can write managed c++ code with little performance drop (though I don't know how much of the.net framework the people writing the test were using).
BTW I still think that system or high performance code should be written in something like C++ but for everything else.net is ok.
If you want bullet proof code then it's going to be boring and a good programmer can always fill in the gaps where older/better tested languages / compilers etc.. don't quite work.
with.net 1.1 I've had things like projects only running in debug mode and crashing when in Release mode. Functions just not working properly, but then they work when you test the code block in isolation e.g. the XML stuff replacing + with a space
encryption routines failing for 1 in about 1000000 calls
Path.IsPathRooted failing randomly.
Most of the bugs went away when I moved everything over to.net 2.0 but then there were a few different bugs in.net 2.0.
I wouldn't trust anything that unstable for any real world application it's just too much of a pain to develop for
When calculators were first invented it may have been a good idea to stick the the abacus until calculators proved themselves, once calculators where proven to be good or there were known work arounds for the issues then more on to calculators.
Just don't use the latest and greatest for critical things use stuff that's been around for a good few years, hasn't had any bug fix patches for a while and is proven.
The point is you shouldn't be running mission critical systems on new and shiney (it's bound to have bugs) you should be running it on old and reliable (or at least where the bugs and workarounds are well known)
I've been on projects time and time again and they keep wanting to use the latest and greatest only to have it break a few months (if their lucky) after deployment.
Assuming it was developed a little while ago they should have stuck to something more robust where people know the bugs and workarounds..net 1.1 was buggy as hell,.net 2 is a lot better but I wouldn't call it mature. maybe in a few years.net will become a platform that you can write more critical systems on but not yet (and that's leaving out the windows side of things)
you can only use the online one if the vechicle is registered to you (I tried!) and because I had just purchased the car it wasn't registered to me and I would have had to have to send the new keeper bit of the logbook off to them and wait for it to arrive and be updated first which would have taken longer than for the insurance to come through, and I wouldn't have that bit to give to the post office to get tax.
Also the last owner was a dealer and drove on trade plates so didn't have any tax at all so there was none to ask for.
I'd already arranged the insurance, but how pray tell do you arrange tax without having purchased the car? I taxed my car within two days since that was the time it took for the cover note to come through.
I purchased a car a few months ago. It didn't have any tax when I got it. I had it parked on the side of the road for 2 days whilst I was waiting for my insurance documents to come through so that I can get tax (it's impossible to get tax without insurance). I was in a catch 22 situation, it was impossible for me to get tax. Anyhow, one of my neighbours dutifully phoned up the DVLA (a government agency) who promptly clamped my car and gave me a £200 fine which I payed promptly. A few weeks later I received another letter from the DVLA this time threatening to fine me £83 for not licensing my vehicle or they were going to take me to court.
I'm going to go to court as I hope that the judge will see that they put me in an impossible situation (but I expect I'll probably end up having to pay an even larger fine)
and that's why most of the middle east are stuck in the middle ages
If you look at countries and philosophies that aren't religious they tend to be no worse or better than the religious ones, except the religious ones are based on the lie of an understanding of God or Gods.
Slashdot Mirror
I second that. On of the things we ask for with candidates is that they submit some code examples before we decide on interviewing them. That usually weeds people out quite well.
As well as the usual questions which only skim the surface and don't really tell you much about ability only about what they can remember. We give applicants some fairly open ended programming excersises and see what they do.
things like factoring some prime numbers and working with roman numerals.
then we have a 3 month period in which to evaluate them further. I like to set them some tasks to see what they are made of and how adaptable they are as well as how well they communicate in a team environment. I usually give them a 'prototype' style application and get them to extend it and make it more extensible for future versions and keeping the same coding style. There are several ways to solve the problem and each have there merits and weaknesses which give something to talk about the task.
Wave/Particle duality isn't fucked up at all. It's kinda quite cosy.
The open university do a few good ones here in the uk (like the Atom series) and the history of maths
shown on the bbc.
They don't go into 'serious' depth and the history of maths didn't actually cover much of the maths details but it did cover the the theroies and the history quite well.
You may be able to find them on a torrent site. (and if your really lucky you may be able to find some of those late night open university broadcasts).
There not really strictly tech but they do cover the science tech is based on.
go lookup linq etc.., there's a lot that Microsoft are doing in .net to make it a lot more parallel.
also you can write managed c++ code with little performance drop (though I don't know how much of the .net framework the people writing the test were using).
BTW I still think that system or high performance code should be written in something like C++ but for everything else .net is ok.
but by the looks of things firefox will have a lot of catching up to do when we start getting more than just a couple of cores in out cpus
if your employes aren't talking then you've got one hell of a productivity risk just in that.
Though most call centres I call are dumb and can't answer questions anyway so I doubt your productivity will be much lower than theirs.
mathematical patents are simply not right for the patent system
There fixed it for you, since all software is just maths.
TextWriters closing the underling steam when they are disposed is just plain crazy
If you want bullet proof code then it's going to be boring and a good programmer can always fill in the gaps where older/better tested languages / compilers etc.. don't quite work.
with .net 1.1 I've had things like projects only running in debug mode and crashing when in Release mode. Functions just not working properly, but then they work when you test the code block in isolation e.g. the XML stuff replacing + with a space
encryption routines failing for 1 in about 1000000 calls
Path.IsPathRooted failing randomly.
Most of the bugs went away when I moved everything over to .net 2.0 but then there were a few different bugs in .net 2.0.
I wouldn't trust anything that unstable for any real world application it's just too much of a pain to develop for
Which version of .net were you using, I've found it to be really buggy.
When calculators were first invented it may have been a good idea to stick the the abacus until calculators proved themselves, once calculators where proven to be good or there were known work arounds for the issues then more on to calculators.
Just don't use the latest and greatest for critical things use stuff that's been around for a good few years, hasn't had any bug fix patches for a while and is proven.
802.11 N supports up to about 300mbps, and has a range of .5 km, wouldn't it be more cost effective to dump a few of these around the place.
The point is you shouldn't be running mission critical systems on new and shiney (it's bound to have bugs) you should be running it on old and reliable (or at least where the bugs and workarounds are well known)
I've been on projects time and time again and they keep wanting to use the latest and greatest only to have it break a few months (if their lucky) after deployment.
Assuming it was developed a little while ago they should have stuck to something more robust where people know the bugs and workarounds. .net 1.1 was buggy as hell, .net 2 is a lot better but I wouldn't call it mature. maybe in a few years .net will become a platform that you can write more critical systems on but not yet (and that's leaving out the windows side of things)
is that legal?
he drove on trade plates so it wasn't illegal.
you can only use the online one if the vechicle is registered to you (I tried!) and because I had just purchased the car it wasn't registered to me and I would have had to have to send the new keeper bit of the logbook off to them and wait for it to arrive and be updated first which would have taken longer than for the insurance to come through, and I wouldn't have that bit to give to the post office to get tax.
Also the last owner was a dealer and drove on trade plates so didn't have any tax at all so there was none to ask for.
I'd also note that I didn't drive the car, I left it exactly where it was dropped off by the person I purchased it from.
kinda short sighted of the people who allowed planning for the housing without adequate off-road parking.
I'd already arranged the insurance, but how pray tell do you arrange tax without having purchased the car? I taxed my car within two days since that was the time it took for the cover note to come through.
I purchased a car a few months ago.
It didn't have any tax when I got it.
I had it parked on the side of the road for 2 days whilst I was waiting for my insurance documents to come through so that I can get tax (it's impossible to get tax without insurance).
I was in a catch 22 situation, it was impossible for me to get tax.
Anyhow, one of my neighbours dutifully phoned up the DVLA (a government agency) who promptly clamped my car and gave me a £200 fine which I payed promptly.
A few weeks later I received another letter from the DVLA this time threatening to fine me £83 for not licensing my vehicle or they were going to take me to court.
I'm going to go to court as I hope that the judge will see that they put me in an impossible situation (but I expect I'll probably end up having to pay an even larger fine)
that's all fine and well, but it sure looks a lot sexier (proper curved surfaces) than rasterization which is what it's competing with
and that's why most of the middle east are stuck in the middle ages
If you look at countries and philosophies that aren't religious they tend to be no worse or better than the religious ones, except the religious ones are based on the lie of an understanding of God or Gods.