Hacker News Reader

Comment by alain_gilbert | Hacker News Reader

Do you usually dig up your garden by hand with a shovel, or do you use a rototiller?

Now we're getting somewhere. This is a great example because obviously speed depends on the context.

Obviously a shovel is quite an improvement for home gardeners but big constructions need a completely different tool.

For +99% of TS projects going from eg 10ms to 1ms is not going to make any difference even though it's a 10x improvement. In part because most monitors refresh at 60hz (16ms per frame) but even with +120hz monitors this is not going to have an impact on any workflow. Even for CI network latency is going to be a much bigger factor than improving TS compilation beyond the performance that TSGO already provides.

I think it might have more meaningful impact on high volume use cases, i.e. where cumulative run time is still significant.

That’s already meaningful for at least a couple of companies I’ve seen investing in these ports. It might be niche for individual dev workflows now, but as those evolve it could become less niche (albeit probably still less uniform).

> I think it might have more meaningful impact on high volume use cases, i.e. where cumulative run time is still significant

For sure but those high volume use cases are extremely rare.

Do you dig your garden with an excavator?

I mean... If I had an excavator it beats using a shovel? This is such a silly "gotcha" analogy because the only reason I wouldn't is cost. And cost is not a reason for tsgo to be slower?

"Would you rather have worse software for literally no benefit?"

We should aim to make software as performant, bug free, feature-complete, and maintainable as is cost effective. It's asinine to argue that software shouldn't be fast, not because of some trade off, but just because someone doesn't need it to be fast for small tasks.

Which, mind you, isn't even the case here. Thousands of businesses use typescript for millions of lines of code. Performance gains can mean tens of thousands of dollars in compute or hardware or time; tons of carbon saved from being emitted.

Yep cost is the reason, allegedly the ts team found go a better language for a port and both using a less ergonomic language and switching target every six months have associated costs

And that's a fine argument to make! But this stupid garden analogy is nonsense.

no, it is the same idea, even if the excavator was magically free there are associated costs like keeping an industrial machine in your garden, danger or heavy machinery and possibly it is going to destroy you garden.

the analogy is not between each specific consequence/cost, it is relative to "single axis optimization" not being a solid strategy

Why are bringing Java into this?

By hand. I've been gardening for years now and I only have a loose idea what a rototiller is.

But in any given year I maybe plant a dozen new plants tops. 6 tomato plants and half a dozen other random things.

Reply on news.ycombinator.com