Hacker News Reader

Comment by ravenstine | Hacker News Reader

I'd bet this has less to do with developer incompetence or frameworks and more with the Peter principle(maybe even the Dilbert principle) inevitably causing orgs to be run by boobs who hand wave about features and redesigns under ridiculous timelines.

I can just picture the board meetings where some poor soul did a presentation of the "new reddit", and one of the bozos saying "That's great, but how will users know that we have a mobile app? Can you add a banner to annoy them into using the app that they don't need? There's a company that promises the greatest analytics in the world ever. You can add their tracking pixel, right? We were wondering about tracking the user's mouse movements. You can do that, right? We hear that JavaScript is the big thing right now. Are you sure we have enough of it?"

DEVELOPER: Sigh.

View on HN

Hmm, I don't disagree a lot of factors affect decisions like this, but... That's not really how board meetings work.

I think it came from a genuine place. You may not agree, and I understand your perspective, but I think there's two things Reddit believes that explain everything you've complained about.

First, that their app is a better mobile experience. And I'd agree, actually, but I understand why someone would disagree. From their perspective it's an easy download and it's native app and it's way better.

Secondly, the best way to build something good is to have more data. This involves tracking people. I think most people have two opposing options: tracking people sucks, and making product decisions in a vacuum is wrong. It's a tough balance.

I think most decisions are done without malice and without "suits" involved.

The breakdown of this generous interpretation to me is that everyone knows people really don't want a ton of apps on their device, specifically for privacy/security reasons. I visit no other site that force feeds their app to the extent reddit does. If they can't set a cookie and fuck off for a long while, forget it. I will never install it because of this, and if ff mobile with desktop requests did not exist, I would no longer visit their site on mobile.

Also what is the point of a native app for a website? Doesn't the browser do a better job already? Are they not just using a webview?

> The breakdown of this generous interpretation to me is that everyone knows people really don't want a ton of apps on their device, specifically for privacy/security reasons.

"But Our Company is different, and that's why we need %useless_feature_they_saw_in_another_app_unrelated_to_industry% to help hook users in."

Admittedly this thinking was a lot more common in the earlier half of the decade, but I've still seen it at other orgs.

There are third-party apps for reddit that are pretty good, including Slide which is GPL and available on F-Droid.

Third party is a bird of a different feather. The point is to modify the site to make it more useable, ala RES. But I have res on ff mobile and that is enough for me.

I use RES with FF on the desktop but on the phone an app like Slide is easily 10x better, IMO.

Firstly, I want to thank you for your politeness. HN can lack basic understanding.

"Board meeting" was probably the wrong choice of words. I have in my career been a part of the kind of meeting I describe, but they weren't actual board meetings, I suppose.

I do agree that Reddit probably thinks their mobile app experience is better. The reason I don't think they have a good excuse is because their mobile site is a prime example of self-sabotage. If they actually wanted people to use it, they would have had just about any redditor tell them that the experience is crap and that it should be fixed. Instead, it comes off like they're intentionally making the web experience bad(or at least neglecting it) in order to convince people to install the mobile app. The way their little popup that they frequently show at the bottom of the screen presents itself even suggests that the native app is the proper way to view Reddit. It's bollocks because Reddit is completely text and image centric; if the web can't do its fundamental job to display those two things, then may God have mercy on our souls.

Don't get me wrong, because I don't think that those in charge at Reddit are doing this things out of ill intent. A bunch of people with good will can convince each other of bad ideas, especially the further removed they are from the front lines of engineering.

I think the only reason they have an app is, that it makes it easier to extract user data. If the app just was “another” choice, without any advantage for reddit, they wouldn’t display a huge and annoying hint to download the app with every page visit.

Many modern webpages are data extraction machines, which is why things got worse in many places. The design is still functional but the main function changed — where it was previously focused on the user and the content it is now clearly all about analytics and ads.

I’d like to say I can’t blame them, because the money needs to come from somewhere, but I am not sure if this is the long term sustainable decision they will praise themselves for when they look back onto things in a decade.

> Secondly, the best way to build something good is to have more data. This involves tracking people. I think most people have two opposing options: tracking people sucks, and making product decisions in a vacuum is wrong. It's a tough balance.

I disagree. In the past decade companies have been collecting more data than ever and software actually became a lot worse than what we used to have. Now it’s all about catering to idiots (the lowest common denominator) and wasting user’s time.

> First, that their app is a better mobile experience. And I'd agree, actually, but I understand why someone would disagree. From their perspective it's an easy download and it's native app and it's way better.

It’s been two years since I last used the native iOS Reddit client, even though I think the responsive web version is embarrassingly bad, because of one reason alone: battery life.

I doubled my screen-on battery life by abandoning the native client.

Maybe in the two year interval they found a way to not be battery vampires but I’m not particularly in the mood to give them the benefit of the doubt.

Keep in mind that just about two literal years ago, the mobile experience was pretty much exactly the same as it is today, if not worse.

I imagine I’m not alone in this perspective: the “smooth, slick, native” app is worse than the stuttering, crashing, mobile web version. Period. Because I value my battery life.

Software has become more and more slower and obnoxious. Everything old, without that magical data is better and faster.

> Secondly, the best way to build something good is to have more data.

The best way to build something good is to have talent, intuition and ideas.

There are people creating wonderful things from a single idea without a minute of market research, and there are people just navigating blindly driven by a bunch of metrics they try to optimise separately.

Usually the latter are mediocre optimizations of mature products, and they often reduce its value for the user.

If your goal is a better product, what do you get out of tracking that you don't get out of just number of clicks? Trying to draw conclusions based on how users maneuver the site seems more dubious with a lot of confounding variables, not to mention creepy.

The app is only better than the mobile website because the website is absolutely terrible. I do read reddit on an app ... but not the official one, I use BaconReader.

One day I suppose Reddit will ban third-party apps for the same reason Twitter did: they're more usable than the official one and feed less surveillance back.

I would be done with reddit the day they ban third party apps, the new site (both mobile and desktop) are insanely slow and tedious to use.

>First, that their app is a better mobile experience. And I'd agree, actually, but I understand why someone would disagree. From their perspective it's an easy download and it's native app and it's way better.

It is still worse than other reddit apps.

>Secondly, the best way to build something good is to have more data. This involves tracking people. I think most people have two opposing options: tracking people sucks, and making product decisions in a vacuum is wrong. It's a tough balance.

There is a difference between analytics that just track usage, and full blown tracking - but nowadays storing data is cheap so why not gather everything and then filter the rest for analysis...

The problem is that analytics, or i should say profiled user data is profitable to sell. Maybe not directly, but allow advertisers to serve ads to very specific users.

> that their app is a better mobile experience

the main purpose of their app is to open third party links. that can't be a better experience in any native app imho

> the best way to build something good is to have more data. This involves tracking people.

This just leads to faster horses. The best way to build something good is to use it

I don't believe that's what the board said. They probably just have a KPI somewhere to increase the share of app usage. Users that have the app are more likely to become long-term reddit users, so increasing that share makes sense. Implementing it in such an annoying way is then someone's decision where the career is dependent on scoring on that KPI.

Some tracking is not going to slow down the site and UI like this. It's just bad architecture and implementation.

Reply on news.ycombinator.com