undefined | Better HN

0 pointsMostAwesomeDude13y ago0 comments

> People’s names are all mapped in Unicode code points. > People have names.

These are apparently not true, so I guess that "a Unicode field in a database" is not the correct answer for my application. So what is? As far as I can tell, patio11's position is that people should simply not gather names, which is an unreasonable position.

0 comments

3 comments · 1 top-level

sp33213y ago· 2 in thread

You don't have to fix every problem for every application. If you don't need someone's name, don't require it (maybe an email address will do). Maybe you only care enough to implement Unicode code points. The problems you actually need to worry about depend on what you need to do with the data in your application.

MostAwesomeDudeOP13y ago

But not everybody has an email address! Oh noes!

No, seriously, this is a bad link, it's got no advice, and it's not helpful.

Here's some serious advice: Put a field in your database, make it Unicode, bound it according to your DBA's advice on how much space you have to spare. A couple thousand code points should be plenty. Then give your users a single field called "Name" and have it accept up to a couple thousand Unicode points. Done.

BerislavLopac13y ago

No, it's not at all helpful in the sense of telling you what you should do. But what it's trying to do is kick you in the butt and make you think for yourself, advising you to discard some of the preconceived notions all of us have about a subject. If it doesn't make you do that about validating names in programming, you're in the wrong profession.

j / k navigate · click thread line to collapse