Hacker News Reader

Comment by Mizza | Hacker News Reader

But I can't get anything at all from the name now. BRCA1 isn't descriptive at all, but suppressBreastTumor is more helpful even it's not 100% accurate. I'd rather it be our best guess description than some random letters.

View on HN

What if we gain more knowledge, and this isn't suppressBreastTumor. What do you do now?

As I see, the answer still cryptic name, and lookup table.

Sure, you'd expect that to happen. There are lots of interesting things you could do there, all of which are better than random letters and numbers that never change. Committees and proposals, objects and references. Things change their names all the time and society is perfectly capable of handling. Pluto isn't even a planet anymore and we all survived that.

I just think we can do better than the first system we came up with, and improving the system would open the whole thing up to more people and possibility.

Isn’t it convenient that you could talk about Pluto without saying “PlutoTheOrbitingObject, which was known as PlutoThePlanet from 1930-2006, isn’t even a planet…”

I mean, we do almost the exact same thing with most celestial objects (ex., HUDF-JD2 aka UDF 033238.7 -274839.8 aka BBG 3179) - but the difference is they don't _do_ anything. I'm never going to write a "program" with planets, but I am interested in writing "programs" using genes. I know it's not a 1-1 example, but you know what I'm getting at.

Are those replacement names or additional names that point to the same underlying object? The former is much more problematic IMO.

BRCA is an acronym for Breast Cancer, so we are already doing that to a large degree.

A fair point, but I think that really just lends credence to what I'm suggesting, that we expand upon that, since it isn't standardized or all that useful. If you were encountering it for the first time, what would you think WntA did?

I would look it up. Genes can encode for multiple things, so it doesn’t make sense to name them like variables, genetics is far far more complex than computer programs. Case in point, the Wnt gene codes for multiple developmental characteristics across insect groups - you can’t simply say they code for this wing protein, it does far more.

Reply on news.ycombinator.com