Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

98% isn't much

whynothugo.nl

Post reply on HN
Print view View original

151–160 of 367 posts

Re: 98% isn't much

#151

Post by AnimalMuppet »

phailhaus wrote: post #115

The broader point is that percentages can be misleading, and are often because of that. It makes things sound better. But usually, the more accurate thing to do is use odds-notation ("1 in 50" instead of 98%). Percentages have a kind of singularity at the edges, where small numerical changes have massive real effects. Going from a success rate of 98% to 99% doesn't sound like much, but that's failing 1 in 50 vs 1 in…

No, you've raised the efficacy from 98 to 99, which is not much of a change. What you have done is halved the inefficacy, which is a big change.

Re: 98% isn't much

#152

Post by steadystate_eng »

phailhaus wrote: post #115

The broader point is that percentages can be misleading, and are often because of that. It makes things sound better. But usually, the more accurate thing to do is use odds-notation ("1 in 50" instead of 98%). Percentages have a kind of singularity at the edges, where small numerical changes have massive real effects. Going from a success rate of 98% to 99% doesn't sound like much, but that's failing 1 in 50 vs 1 in…

[flagged]

Re: 98% isn't much

#154

Post by robalni »

I don't like treating people like numbers. 98% isn't much and it isn't little. It's just wrong.

If I'm one of the 2% then that's everything for me. Maybe I have good reasons to be in the 2%. And maybe, not caring about that is wrong.

I would rather have a website that only works for 2% of people for the right reason, than a website that doesn't work for 2% for the wrong reason.

Re: 98% isn't much

#155

Post by maderalabs »

I think there are broadly two types of problems - ones where you get partial credit, and ones where you don't. The restaurant example is one where you don't get partial credit - 98% of food being safe isn't enough, it's all or nothing. Paying your employees - all or nothing, you miss a paycheck once, it's a huge problem.

CSS on a website, however, you CAN get partial credit (to an extent). It may not be perfect, but it's at least theoretically still providing some value partially.

I think knowing what kind of problem you're facing is really important when it comes to measuring percentage of "complete".

Re: 98% isn't much

#156

Post by ben_w »

londons_explore wrote: post #72
bradleybuda wrote: post #42

Earlier quoted context omitted.

It’s very difficult for the average person to use a ten year old browser; in fact I’d offer that the only way to use a ten year old browser is to be an expert and do so intentionally.

There are plenty of people with old android phones with no free disk space using ancient browsers. There are plenty of people still using windows 10 with updates turned off or wedged for whatever reason. These people just use the sites that work. They aren't computer experts, and might not even realise why half the internet doesn't work - they just think that's the way things are.

While I agree with your general gist and definitely your final paragraph,

> There are plenty of people with old android phones with no free disk space using ancient browsers.

How many people have 10 year old phones? I've got an 8 year old iPhone XR which I keep around as a backup/travel device because it's not worth selling, and the battery is… not happy even in airplane mode.

For me to have a 10 year old mobile browser, I'd have to have kept the iPhone SE 1 (or was it a 5c?) that I bought second hand in 2018, and not upgraded it since I bought it. I got rid of it because the battery wouldn't hold a charge for 10 minutes.

Re: 98% isn't much

#157

Post by igsomething »

account42 wrote: post #138
igsomething wrote: post #119

Earlier quoted context omitted.

For public services you can tell people to use another device, or provide a way to schedule an appointment in-person that is accessible using old browsers.

You you can make a law that requires such businesses to use perfectly good technology standards that are widely supported instead of whatever EEE crap the latest Chrome comes with.

I agree, but one thing is to demand all your users to be on the latest Chrome, and another one is to support browsers that are no longer maintained and contain security issues (IE). If we discourage people from driving old insecure cars, we can also discourage people from using old insecure browsers.

Re: 98% isn't much

#158

Post by theresistor »

I had this argument with people working on VR headsets, where a physical parameter was designed to cover the 5th to 95th percentile. I had to point out that flat-out excluding 10% of the population is a pretty crappy starting point...

Re: 98% isn't much

#160

Post by johannes1234321 »

One thing I wonder regarding browser market share is always: How is it collected?

I assume Firefox users over proportionally use privacy extensions.Thus they overproportionally won't appear on Google Analytics and similar places, which for some statistics reduces the numbers even more than reality.

Post reply on HN