The quals don't matter - what you enjoy does
on
CS vs CIS
·
· Score: 1
First off, let me apologise that being from the UK I don't really understand all your education based acronymns. However I am sure the principles are the same.
I have worked for what is now a major systems company for 30 years. We take on a few hundred graduates every year, and in the UK, we generally get our pick of the cream, given our reputation.
At a first approximation, the actual details of the degree matter nothing - obvously a degree in a computing subject is better than a numerate degree. A numerate degree is better than an arts degree and we pay a few hundred pounds a year more to those with qualifications above degree level. But very quickly (ie within a few months) it is your performance in the job that counts. Are you taking reponsibility for your work, are you helping the company by helping understand and solve its clients problems, are you showing leadership skills, are you able to coach others, are you able to work well in a team and are you able to network within the company to find those who can help you solve problems when you can't. Of course pure technical skill is important too - but its more an inhibitor if you haven't got it or can't pick it up quickly when you need to use them. The other attributes enable to you progress much faster.
Most of this means that to be able to do these things well you must enjoy working. Therefore - take a course that interests you, learn to do the things you like doing well and not because you think they will ultimately get you more money.
A great set of requirements with which I would almost entirely agree. Here are a couple more musts for me
10) Synchonise with PDA (in my case Palm Pilot) locally (I also run a laptop)
11) Allow my secretary to arrange meetings on my behalf (she will typically be doing this between a group of about 5 of us - it needs to be dead easy for her to manage multiple calanders at the same time)
12) Ability to manage time zone changes gracefully. (let me explain). If I am on a trip to the US from the UK, I switch to US time in my mind, but I don't go into windows and change my time zone. Nevertheless I want (on my laptop and my PDA) to see meetings at the time in my mind. My secretary back in the UK must also see my calender and understand both the "real" time (in case these meetings are teleconferences with people in other parts of the world) and my "mind" time (so she doesn't schedule a meeting when I would rather be asleep)
Clearly reality will prevent us actually building a device like this so it will be a few orders of magnitude less powerful. However the article concluded that such a device could be built within the next 200 years (following Moore's law)
That doesn't seem very far away given I can remember computing 40years ago.
And Australia in particular has massive tax breaks for doing R&D in their country. Much of this could be "free" to IBM as the hardware that gets shipped into Australia is freed from import duties as a result
The company I work for has just put out a press release that shows it is about to be developed for the European and Japanese Mobile Phone Market.
Now if that gets going then there really will be a micropayments system of interest.
Clearly the price of food of all sorts drops through the floor
However - there is still a market for someone to think through what a balanced diet for the week ought to be, match it to my lifestyle (individually taylored to the way I like it) and delivered to my front door. They may even offer a self selection service as part of this. I will pay, not for the physical fruit, but (because I can now afford it) the variety and the service.
In order to provide this variety, they will need to pay the farmer to invent new varieties of fresh fruit to include within their portfolio and will pay him sufficiently well to keep him in decent living conditions just to grow one item of fruit, just as long as he comes up with one new variety every few months.
If the farmer is enterprising, he will probably sell tours round his farm which will have a different variety of fruit on each bush/tree and will invite you to feel and touch and squash each different variety to compare one with another. It won't of course matter how much fruit to touch and squash and occassionally taste.
Everybody wins who adapts to this new situation, although the farmers who still try to grow single varieties, or shops that still think they can sell you goods without the addition of useful side services will go out of business. They will of course kick, scream blue murder and try to get those in power to ban the new replicator machines as they go under - but go under they eventually will unless they adapt.
The real issue with most reviews (and I exclude those that have done detailed performance evaluations and are reporting on the results) is that there is just not enough time to do them. Most reputable companies are putting out products that you could not expect to find holes in without some prolonged use or a in set of very non standard circumstances. They can after all be expected to have done some decent testing before they ship.
Why then would one expect the majority of reviews with at most a couple of weeks trying out the software expect to come up with anything positive.
What really tests this sort of thing is actual experience in actual production environments.
That said, I don't think that reviews should be ignored. For me, at least, they have two uses.
Firstly, they are a useful prefilter, not on the quality of the product, but on the suitability of the product for your particular need. This is a often a helpful pointer before further analysis
Secondly, they fulfil an important emotional need. I often read reviews of software after I have bought it to get a sort of justification for the money I have spent. In these circumstances I don't want to be told I have made a mistake:-(
Why bog down the processor with a full embedded operating system? I don't understand this. I design embedded systems for a living and I'd say a full 90% of the systems out there have absolutely no need for a full (or even stripped down) Linux (or anything else) kernel. Cellphones?! gimme a break! Because the next generation of these devices will be full blown Internet Capable, computing devices. With 300K+ comms to and from these things and the need to recoup $billions in investments some power and flexibility will be needed.
First quality needs to be endorsed from the top of the company to create a culture in which it is encouraged.
Secondly, there are other alternatives to firing people - training them (either through mentoring from the more senior people, or via formal training courses) can also sort out problems.
Thirdly, quality, as the book says cannot be chrome plated on the outside, but requires it to be built in to the low level of everyones activities. This requires everyone to take pride in what they do - and this in turn requires the organisation (and in particular the management) to recognise the value in everyone (well most people - there are always one or two that cannot or won't be redeemed:-) ).
But as you said, there arent enough hours in the day to do all the exta stuff. It costs extra to do this work, especially in the same time the work can be done with out it.
It might cost extra to you the individual, but a good approach to software engineering saves time and effort to the team and for that reason it is cost effective even in the short term for any activity that is more than one person to have a documented approach to this. On any but the most trivial of project, accurate communication of the requirements and the technical interfaces is what dominates costs (mainly in fixing misunderstandings during integration testing).
However, you have to be careful on methodologies. There is no silver bullet and you have to design the set of methods that you use to fit the circumstances. Inappropriate methodologies do do more harm than good.
I spent much of the 1980's trying to prevent our company from imposing such inappropriate methodologies on all development projects (we are a computer systems house and this is our bread and butter). Our approach these days is to build a [we call it] 'Quality Plan' from a set of predefined standard bricks (we try to envisage most environments and have developed over 30years a whole set of standards that are appropriate to each of the situations we encounter, we have ongoing feedback as new technology comes out to try and improve on these) - but with the option in any particular circumstance to include a customised version of any standard.
Slashdot Mirror
First off, let me apologise that being from the UK I don't really understand all your education based acronymns. However I am sure the principles are the same.
I have worked for what is now a major systems company for 30 years. We take on a few hundred graduates every year, and in the UK, we generally get our pick of the cream, given our reputation.
At a first approximation, the actual details of the degree matter nothing - obvously a degree in a computing subject is better than a numerate degree. A numerate degree is better than an arts degree and we pay a few hundred pounds a year more to those with qualifications above degree level. But very quickly (ie within a few months) it is your performance in the job that counts. Are you taking reponsibility for your work, are you helping the company by helping understand and solve its clients problems, are you showing leadership skills, are you able to coach others, are you able to work well in a team and are you able to network within the company to find those who can help you solve problems when you can't. Of course pure technical skill is important too - but its more an inhibitor if you haven't got it or can't pick it up quickly when you need to use them. The other attributes enable to you progress much faster.
Most of this means that to be able to do these things well you must enjoy working. Therefore - take a course that interests you, learn to do the things you like doing well and not because you think they will ultimately get you more money.
A great set of requirements with which I would almost entirely agree. Here are a couple more musts for me 10) Synchonise with PDA (in my case Palm Pilot) locally (I also run a laptop) 11) Allow my secretary to arrange meetings on my behalf (she will typically be doing this between a group of about 5 of us - it needs to be dead easy for her to manage multiple calanders at the same time) 12) Ability to manage time zone changes gracefully. (let me explain). If I am on a trip to the US from the UK, I switch to US time in my mind, but I don't go into windows and change my time zone. Nevertheless I want (on my laptop and my PDA) to see meetings at the time in my mind. My secretary back in the UK must also see my calender and understand both the "real" time (in case these meetings are teleconferences with people in other parts of the world) and my "mind" time (so she doesn't schedule a meeting when I would rather be asleep)
Sometimes yes - but there should be an option for speed where that transparency is lost.
Clearly reality will prevent us actually building a device like this so it will be a few orders of magnitude less powerful. However the article concluded that such a device could be built within the next 200 years (following Moore's law) That doesn't seem very far away given I can remember computing 40years ago.
And Australia in particular has massive tax breaks for doing R&D in their country. Much of this could be "free" to IBM as the hardware that gets shipped into Australia is freed from import duties as a result
The company I work for has just put out a press release that shows it is about to be developed for the European and Japanese Mobile Phone Market. Now if that gets going then there really will be a micropayments system of interest.
surely it should be peeking order:-)
Clearly the price of food of all sorts drops through the floor
However - there is still a market for someone to think through what a balanced diet for the week ought to be, match it to my lifestyle (individually taylored to the way I like it) and delivered to my front door. They may even offer a self selection service as part of this. I will pay, not for the physical fruit, but (because I can now afford it) the variety and the service.
In order to provide this variety, they will need to pay the farmer to invent new varieties of fresh fruit to include within their portfolio and will pay him sufficiently well to keep him in decent living conditions just to grow one item of fruit, just as long as he comes up with one new variety every few months.
If the farmer is enterprising, he will probably sell tours round his farm which will have a different variety of fruit on each bush/tree and will invite you to feel and touch and squash each different variety to compare one with another. It won't of course matter how much fruit to touch and squash and occassionally taste.
Everybody wins who adapts to this new situation, although the farmers who still try to grow single varieties, or shops that still think they can sell you goods without the addition of useful side services will go out of business. They will of course kick, scream blue murder and try to get those in power to ban the new replicator machines as they go under - but go under they eventually will unless they adapt.
The real issue with most reviews (and I exclude those that have done detailed performance evaluations and are reporting on the results) is that there is just not enough time to do them. Most reputable companies are putting out products that you could not expect to find holes in without some prolonged use or a in set of very non standard circumstances. They can after all be expected to have done some decent testing before they ship.
Why then would one expect the majority of reviews with at most a couple of weeks trying out the software expect to come up with anything positive.
What really tests this sort of thing is actual experience in actual production environments.
That said, I don't think that reviews should be ignored. For me, at least, they have two uses.
Firstly, they are a useful prefilter, not on the quality of the product, but on the suitability of the product for your particular need. This is a often a helpful pointer before further analysis
Secondly, they fulfil an important emotional need. I often read reviews of software after I have bought it to get a sort of justification for the money I have spent. In these circumstances I don't want to be told I have made a mistake:-(
Why bog down the processor with a full embedded operating system? I don't understand this. I design embedded systems for a living and I'd say a full 90% of the systems out there have absolutely no need for a full (or even stripped down) Linux (or anything else) kernel. Cellphones?! gimme a break! Because the next generation of these devices will be full blown Internet Capable, computing devices. With 300K+ comms to and from these things and the need to recoup $billions in investments some power and flexibility will be needed.
But this breaks the rule that says you cannot register just the TLD
First quality needs to be endorsed from the top of the company to create a culture in which it is encouraged.
Secondly, there are other alternatives to firing people - training them (either through mentoring from the more senior people, or via formal training courses) can also sort out problems.
Thirdly, quality, as the book says cannot be chrome plated on the outside, but requires it to be built in to the low level of everyones activities. This requires everyone to take pride in what they do - and this in turn requires the organisation (and in particular the management) to recognise the value in everyone (well most people - there are always one or two that cannot or won't be redeemed:-) ).
It might cost extra to you the individual, but a good approach to software engineering saves time and effort to the team and for that reason it is cost effective even in the short term for any activity that is more than one person to have a documented approach to this. On any but the most trivial of project, accurate communication of the requirements and the technical interfaces is what dominates costs (mainly in fixing misunderstandings during integration testing).
However, you have to be careful on methodologies. There is no silver bullet and you have to design the set of methods that you use to fit the circumstances. Inappropriate methodologies do do more harm than good.
I spent much of the 1980's trying to prevent our company from imposing such inappropriate methodologies on all development projects (we are a computer systems house and this is our bread and butter). Our approach these days is to build a [we call it] 'Quality Plan' from a set of predefined standard bricks (we try to envisage most environments and have developed over 30years a whole set of standards that are appropriate to each of the situations we encounter, we have ongoing feedback as new technology comes out to try and improve on these) - but with the option in any particular circumstance to include a customised version of any standard.
I wanted to get the updated packet filtering functionality of iptables (as opposed to ipchains and ipmasqadm portfw).
I upgraded and ppp crashed. Needed to get the new release of that.
Also had some problems with where the modules had been put to get depmod (etc) to find them.