4 ms·

Count me among those who who thinks the CS vs. Software engineering debate is silly. I never understood how you could teach CS theory w/o a lot of programming

by uuilly 18y ago

Count me among those who who thinks the CS vs. Software engineering debate is silly. I never understood how you could teach CS theory w/o a lot of programming and I have no idea how you'd teach programming w/o the theory. If you can't figure out the big-O complexity of an algorithm you're useless and if you can't implement it you're useless. I also think that engineering degrees should not be crammed into 4 years. Just to get the basics requires way more credits than a normal BA. Perhaps a BS should be viewed as something between a BA and an MS and 5 years should be standard.

gaius 18y ago
Because most people working in software development aren't doing anything remotely algorithmic. The vast majority of corporate/commercial software does two things: it gets input from somewhere, either a user or another program, and writes it to a database. It gets records from the database and formats a report for a user or sends it to another program. And that's it. That's all most websites do too. If it's XML then the programmer doesn't even need to know parsing. You can - and many people do - work in software and be good at your job, without any theoretical underpinnings at all. That's not to say that that's a good thing, mind, it's neither good nor bad, but it's a fact that maybe 1 in 10 people employed in the software industry will need to know big-O, perhaps even less than that.
- azanar 18y ago
  That's not to say that that's a good thing, mind, it's neither good nor bad, but it's a fact that maybe 1 in 10 people employed in the software industry will need to know big-O, perhaps even less than that. I don't think the ratio is that high, simply because most people who do software development aren't in the software development industry. They are members of an IT team in an organization who profits come from the sales of things other than software. Although the organization they are at might benefit from software that solves real technical problems relating to their product line, most of the time, they are just writing ETL programs that require little more than getting table names right. Of the people who work at companies that see software development as one of their core disciplines, I imagine the problems are sufficiently hard to make some knowledge of theory useful.

by uuilly 18y ago

gaius 18y ago

Because most people working in software development aren't doing anything remotely algorithmic. The vast majority of corporate/commercial software does two things: it gets input from somewhere, either a user or another program, and writes it to a database. It gets records from the database and formats a report for a user or sends it to another program. And that's it. That's all most websites do too. If it's XML then the programmer doesn't even need to know parsing. You can - and many people do - work in software and be good at your job, without any theoretical underpinnings at all. That's not to say that that's a good thing, mind, it's neither good nor bad, but it's a fact that maybe 1 in 10 people employed in the software industry will need to know big-O, perhaps even less than that.

azanar 18y ago
That's not to say that that's a good thing, mind, it's neither good nor bad, but it's a fact that maybe 1 in 10 people employed in the software industry will need to know big-O, perhaps even less than that. I don't think the ratio is that high, simply because most people who do software development aren't in the software development industry. They are members of an IT team in an organization who profits come from the sales of things other than software. Although the organization they are at might benefit from software that solves real technical problems relating to their product line, most of the time, they are just writing ETL programs that require little more than getting table names right. Of the people who work at companies that see software development as one of their core disciplines, I imagine the problems are sufficiently hard to make some knowledge of theory useful.