It is pretty offensive that they say "reddit works better on the app" when the only reason for that is that they broke everything on mobile (on purpose?) in a series of badly-implemented redesigns.
I'm never going to install it, and I have all but stopped reading it because these prompts are so obnoxious. That's probably +$ for Reddit though.
I don't understand why companies constantly do stuff that serves only _themselves_, and then expect users to engage with it because it exists. Users are able to identify when something is valuable to them. If you make it valuable they will use it. Consider the difference on an ecommerce site between a comment section vs a few company-picked "testimonials" above the fold. _Everyone_ knows the testimonials are garbage. Maybe your conversion goes up a tick the first time you put them on the site, but when a repeat viewer sees the same ones again they're going to roll their eyes and register you as untrustworthy. Whereas a (reasonably-managed, honest) comment section provides loads of information that's actually valuable to the consumer.
MIT economists Acemoglu and Robinson wrote a book "Why Nations Fail" that analyzes why countries at similar starting points can have extremely divergent paths in terms of economic development. One example is South and North Korea. Their conclusion is that some countries have established extractive institutions, while others inclusive. The extractive ones exclude the population from the decision making process and seek to divvy up and redistribute the scarce available resources, while the inclusive ones seek to involve the broadest possible strata of the population in the decision making and wealth-producing activities. Looks like Reddit is turning into the North Korea of social websites.
The extractive rent-seeking (slavery being the most extractive economic and political institution possible) South was significantly lagging even before the rebellion, which was crushed by the United States whose other regions had more inclusive economic institutions.
In more recent times the authors compare two towns across the US-Mexico border: Nogales, AZ to to Nogales, Mexico. Even thought the language and ethnicity is mostly the same (95% Hispanic), with same climate and what not - Nogales in the United States has three times the GDP per capita of Nogales, Sonora. The difference is more inclusive economic and political institutions in the United States - while Mexico has suffered under one party rule for decades.
Which has been the case since almost the inception of the country, as it is a direct consequence of the First-past-the-post voting system that the US government has implemented.
You're correct, and I've love to have something other than first-past-the-post.
However, throughout our history, our two-party system has generally (with some notable exceptions) been able to work and work together. That's no longer the case. As such, it's hard to tell if the political polarization is a reflection of society, or if society is just adopting the polarization of our political elites.
This is a tremulously rose colored portrait of history. The most openly hostile, contentious election in American history didn't happen in 2016, or 2008, or 2000, it happened in 1828. Many American institutional families would say that Jackson being a jackbooted thug was the very least damning of his odious traits. Meanwhile Jackson's largely populist supporters would go on to accuse the current poltical body of being wholly corrupt, and hellbent on creating a new feudal aristocracy, where the people's wills and desires were controlled through the national bank. Honestly 2016 just reads like a rerun of 1828, except Clinton wasn't publishing articles about how Trump was likely to gun her down in cold blood if he lost.
I suggest reading more carefully. I wrote “with some notable exceptions” very deliberately. I had exactly the Jackson presidency and the 1850s-1860s in mind.
But I figured that HN is an educated audience, so no need to specify the few exceptions I had in mind.
BTW, 2008 and 2000 don’t even compare with 1828, or 2016. The election of 2016, and the (only partially merited) furor that has followed it will unquestionably stand out in US history, quite possibly as a prelude to our descent into another civil conflict.
And what’s rose-colored is making an oblique claim that our current political situation is nothing more than business as usual.
While I think a slow trend of increasing contention exists, I don't buy that hostilities were really so much better in '08 than in '16, and I think we've a ways to go yet before we starting debating whether capping our political enemies is a barbaric but necessary practice.
Ironically, I believe it's the increase in democratic processes in American politics that's responsible for this.
See, historically, when we spoke about parties, we really spoke about their respective establishments. The voters decided which one would be running the show at any given moment, but internal party politics were not readily accessible to the outsiders, and thus candidates and platforms were also vetted by the party establishment. Since establishment tends to be conservative by its very nature, this provided a screen against more extreme positions. To be clear: the politicians would still run on extreme positions, if they saw it as advantageous - but that didn't mean they'd govern accordingly if they won.
Then we got primaries. It's not that they didn't exist before - the first ones date back to before the Civil War. But they weren't anywhere near as common until the second half of the 20th century, and even then most were non-binding (and their results were often ignored when the "wrong" candidate won them).
And the primaries themselves changed over time, generally so as to broaden eligibility. In most other countries (where primaries are internal party matters), you have to actually be a member of the party in good standing in order to participate in that party's primary. The American system, where you merely need to register party affiliation in advance, is much more lax - and by now, even that formality is no longer necessary in many states.
So, voters now have a lot more influence over party politics. And this caused a nasty feedback loop, where agitprop that's thrown by the establishment at the voters, solely for the purpose of convincing them to vote one way or the other (appealing to emotions, especially fear and anger, since it's the most effective way), is regurgitated and injected into internal party matters and policies. So the party platform moves further away from the center, and its agitprop shifts accordingly, creating even more radicalized supporters, who again feed it back into the party etc.
Tea Party was a wake-up call in that sense, when a bunch of traditional Republican candidates lost to grassroots crazies who said mostly the same things, but who also sincerely believed in those things. Trump is also part of the same phenomenon - he might not be a sincere believer, but he appears as one to his followers. And there's some evidence of similar processes happening in the Democratic party. Slower, because its electorate is a broader coalition with more conflicting goals, and so the direction in which the voters move the party is less clearly defined. But you can see the internal radicalization spill out on some issues already, and there's more to come.
It has nothing to do with first-past-the-post. You can have 10 parties, 100 parties, in fact first past the post is designed for more than two parties.
If you have people in 5 parties to vote for in your riding(district?), then first past the post is the only way it works easily! Seldom will one of those 5, get 50.1% of the vote!
And why would they? The most popular wins, the people have spoken.
I... think you may have some reading to do. The subject is pretty well studied.
The main reason FPTP ensures at most 2 parties is that, say you have parties 1, A, and B, where A and B are relatively similar to each other. And say A and B are together more populous than 1, like a 40/30/30 split.
By splitting those 60% mostly-similar votes between A and B, both lose to the less-popular 1. It becomes in A and B's best interest to not cannibalize each other's votes, i.e. merge and dominate (or at least compete). The less fragmented ideology wins, not the most desired.
It's even more compelling during the formation of new parties when there are only 2: by not voting for one of the two dominant ones, you are literally throwing your vote away because it has no chance, and it's worse for your ideology than if you had chosen the most-similar of the dominant party.
Canada has both Federal, and Provincial FPTP parliaments. On the Provincial level, new parties appear, and are elected , minority or majority, all the time. Constantly.
On the Federal level, things move more slowly. Yet new parties are appear, grow in popularity, and replace older parties.
Canada is not only a place of multiple parties, it is a place of constant party renewal.
And as a Canuck, I am all too familiar with vote splitting. We have party mergers. Parties that split and form new parties.
We know all about strategic voting, but because we have the concept of minority governments, and because we don't vote for who our Prime Minister is, the dynamic changes a bit.
FPTP isn't what causes two parties only, it is certain methods of government that do, mayhap combined with FPTP.
It seems the split President (prime minister) and House voting leads to more issues compared to Canadas system. Canada also has no confidence votes vs fixed elections ...
That makes sense if you think in terms of parties. But if you're interested in policies instead, I think it might be wrong.
For example, if you want them to do something about climate change you might vote for the Green party instead of Democrat. They won't win but your "stolen" vote might cause the Democrat party to lose. They don't want to lose so they might adopt policies from the Green party to attract you and all Green voters. That would be as good an outcome as multiple parties.
Canada uses FPTP, and yet there are five parties represented in the house of commons [0].
While it's true that two parties have the lion's share of seats (155 for the liberals, 121 for the conservatives), as there are 338 seats, the liberals can't just steamroll legislation unopposed; they have to get at least some approval from other parties.
But yes, Canada has soldiered on remarkably well in spite of it. USA used to be multi-party as well, the transition to 2-party takes time and there's nothing that truly guarantees it, just innately encourages. There are ways to slow it down / isolate sections... but the dominant parties can also whittle away at those over time.
It should be noted that the merger between the older PC party, and the reform/alliance party was with the Alliance party at massive strength, in the House of Commons with a respectable number of seats, and 100% in charge of the future of the merged party.
My point here is ; the willingness for people to embrace new parties, new ideas, and vote for people .. not parties, helps with real political renewal in Canada.
Further, while some provinces aren't, most are MORE dynamic than the federal level at party change. Take Quebec, which has had new parties appear and then even form governments, in a period of two election cycles.
https://en.wikipedia.org/wiki/Canadian_electoral_system#Firs...
> An absolute majority of the electorate is not needed, and is rarely achieved. As a result, power has been held by either of two parties for most of Canada's history. The party whose candidates win the second largest number of seats becomes the Official Opposition.
It doesn't ensure at most two parties. The UK has at least four major parties: Conservatives, Labour, Lib Dems, SNP, and then one or two other smaller parties.
Districts are drawn more or less arbitrarily, and so their residents don't necessarily represent a single coherent political bloc. In practice, this means that you can have a party getting 30-40% of the vote in literally every district, and end up with zero representation.
Worse yet, because most FPTP elections only require plurality for victory, you can have a party get less than 50% of the vote in each district (but more than any other party), and still take all the seats. In which case you literally have the minority running the country - so much for "people have spoken".
>first past the post is the only way it works easily
The key word there is easily. Now that we have computers and instant communication we can do things that would be/used to be considered hard but give better outcomes.
I'll also point out you don't need a winning party in an election you just need to be able to form a government and there's no reason 2 closely aligned parties can't work together or even traditionally opposing parties.
Why does someone need a magic number of 50.1% to be valid?
Why is that "the number"? Because it's more than 1/2? You're thinking "two parties".
If you're worried about "most people's support", then why isn't the number 80%? 99%? 100%? Why does '50.1%' make it 'OK' and less 'not'?
I don't get this arbitrary fixation on "but, that guy didn't get 50.1% of the vote!!". So?
Let's say there are 5 parties. One gets 40% of the vote. Yet if each party had equal votes, that would be 20% each. The party with 40% of the vote, received 2x the votes compared to the number of parties.
Is that OK? Why not?
I don't get what's wrong with a number less than 50%. I don't get why anything above 50% is fine.
Let's say you have 10 people, and 4 plans. Plans to defend the city from invasion. Each plan has its merits. You ask people to vote. 2 abstain. The next 2 receive two votes each. The last receives 4 votes.
4/10 is less than 50%, yes? Yet, clearly it is the most popular. Must the 2 abstainers vote? Must those that don't agree, be forced to vote for a plan they don't agree with?
How is this undemocratic? Were the people's voices heard? Was the most popular plan voted on?
I literally don't get it, or understand the issue.
Indeed. This seems to be a natural progression of yin-yang where these processes will unfold as part of an education process, and then once policy is implemented to counter the two party system then I can only imagine society beginning to thrive immensely relative to before; ranked choice voting, $100/year budgeted for each eligible voter to contribute to politician of their choice, etc - a few ideas Andrew Yang was running on.
Is this a real book someone wrote? Incredible to me how you can call yourself an economist and just write basic observations and conclude they are the root cause of everything without even taking a cursory look at the history of things.
It's kind of like how you can read a post on the internet that mentions two economists and draw any kind of conclusions about the work they've done without even taking a cursory look at it.
I would imagine the book contains a bit more justification for its arguments than what was presented by the person you are responding to.
What makes you believe that the author just "wrote basic observations and concluded they are the root cause of everything"? Most ideas are easy to dismiss when you assume that a blurb posted to HN is all there is to them.
The most enraging thing is Reddit took over video hosting themselves, but simply declined to have a mobile/text share option.
Links from any 3rd party service show up in the phone as the content i want to share.
Reddit? You try to send someone a Reddit video link and they get affronted by a pop up banner to download the mobile app when they click through. WTF!
I sent a reddit link to a video to my dad, once. He replied 'cant open it.' So f* off reddit. Allow rehosting or change your content policies. Stop making the internet worse.
On top of this, v.reddit.com is abysmal. Videos stop streaming, the cursor is impossible to use, I don't know how its allowed to exist in its current state.
Videos show the wrong duration, sometimes, as well.
A couple of weeks ago, someone was asking for the full video of what looked like a 10s clip (which also played just 10s). Someone else replied, that it was literally the post. He got downvoted to hell (like below 100). He was right. After checking his comment, I went to the video and by seeking I got it to show the whole 2m and something.
This was not the first time this issue has appeared for me.
What's frustrating is that many subs will ban the bot, so getting the mirror link requires actually private messaging it. Every once in a while someone who did this will reply to the empty in-thread request with the download / mirror link though.
v.reddit videos are nearly impossible to watch on the browser version of Reddit. If you download direct you don't get audio because they stream video and audio independently (I Think). They removed the shortlink to the hosted site on the browser version too. They're really doing a lot of user hostile things for monetization.
Furthermore, over the long term it can change the texture of the Reddit community.
What sort of person quits using Reddit after becoming frustrated with yet another imposed hassle?
What sort of person stays, or is retained by pop-up notifications and whatever other tedious blandishments their app/redesign provides?
There are other sites with similar features and layout to Reddit (Voat, Ruqqus, Saidit, Raddle, communities.win, etc.) but none are really in competition to be "the front page of the Internet".
This is because Reddit has a huge community with varied interests which provides self-perpetuating advantages.
For example: the activity level is generally higher; if you start a subreddit for a niche topic you have a chance at finding an audience; great comments/submissions will garner thousands of upvotes rather than a few dozen (smaller sites have a proportionately tinier vote ceiling, making it hard to differentiate between "good" and "amazing").
I can see increased retention/monetization maybe leading to a larger user-base in the short term, but if it makes good content less likely to crop up, or leads to less discerning users (ingenious effortposts lose out to low-effort pandering memes), or even if it just narrows the interests/hobbies represented on there, then over time it will destroy Reddit's main source of value.
I'm sure good subreddits still exists, but it has gone way past eternal September at this point and Gresham's law is in full effect. Once an organization reaches a certain size, vested interests on all sides have perverse incentives, and it starts a terminal slide to the lowest common denominator. I suspect the issue is largely in the inherent zero marginal cost to entry for most social media platforms.
Ruqqus is generally better as an alternative and has been doing a better job navigating the line between free speech and not letting neo nazis invade every sub.
Like old school reddit, people do say inflammatory things to get reactions but usually get downvoted. There are also good tools for muting people.
Agreed. I tried consistently visiting and posting for a week or two,and while I feel like the site has been getting more popular, it's not a place for adult conversations.
What finally made me quit was a post of a meme involving Kamala Harris in a non-political section which was one of the most upvoted things I'd seen on the site, yet the meme itself was childish and racist.
I dislike censorship, but if the initial source of people visiting a site which has these anti moderation policies are low quality, good luck climbing out of that hole. Smart people are attracted to smart people, so your initial userbase is paramount. This isn't a condemnation of all on ruqqus, but the majority are looking for a right wing echo chamber.
The site is nowhere near as bad as voat, but give it a little time, it'll get there.
But it'll only avoid becoming voat if people invest time to grow the community into a place of meaningful discussion and beneficial discourse. There are some sub mods who are trying very hard to build welcoming (and non-political) subs but they can't do it singlehandedly.
If you want something to grow, you have to invest in it.
I don't know why anyone would bother with the first party app when there are a bunch of really nice 3rd party ones with all sorts of additional features. All the one's I've used have been free with premium versions that only cost a few bucks.
Which of course leaves me wondering how long it is before reddit pulls the plug on their API and forces people to use the busted mobile site or their app.
I hope they don't do it. If this happens I probably won't Reddit anymore. Apollo is not only the best Reddit client I've ever used, but is actually the most well-crafted iOS app I ever had the pleasure to use.
>Which of course leaves me wondering how long it is before reddit pulls the plug on their API and forces people to use the busted mobile site or their app.
Of course. How will they do it? That's yet to be seen. Instead of pulling the API entirely, I'd bet they simply wont introduce API for new features. This will leave 3rd party apps as inferior while Reddit doesn't have to deal with the backlash of pulling the API entirely. I'd argue this is already occurring.
> (...) and then expect users to engage with it because it exists.
This is indeed a troubling trend, so many annoying popups, banners, and notifications only seem to serve to forcefully try and push a small percentage of users towards some feature. If adoption is poor, it's probably because it's not wanted. Of course, in some SaaS contexts, it makes a lot of sense to keep users informed of features that might be of benefit to them and they may have otherwise not discovered, but that's a completely different game from just shoving stuff down their throats.
What's even more frustrating is the dark UX design they implemented. If you make it past their "Continue" pop-up which shows you multiple versions with the options reversed (to work against impulse memory skips), if you touch ANYTHING on the page after that point, it will STILL open up in the app.
What's worse is that once it opens, that post is now permanently in your account's history. Reddit provides NO WAY to remove things from your account history.
Whenever any website says "works better in the app" I always roll my eyes and consider it to be really saying "we really just want to spy on your data (apps installed, wifi networks, geolocation, accelerometer to determine your life habits, ...)"
For a site like reddit there is literally nothing they can't do with pure HTML/CSS/JavaScript.
Also, I like Firefox on mobile because you can block ads and install other extensions.
In a similar vein, I use Apollo for iOS[1] and it's also magnitudes better. The official reddit app is starting to catch up, but it's designed with advertisers in mind, whereas you can use a different app and pay for ads to go away, or use the old, web version with a blocker.
Apollo is among the best. The developer Christian is super engaged with his user base, and the local animal shelter. He has donated $10,000s to them through fundraising!
Take heart, they're actually doing you a favour. Once it becomes a hassle, you'll stop visiting, and along the way you'll wonder why you ever did. We should be thanking them for helping break the cycle of looking at a bottomless pit of drivel, our time on earth is not without end.
Anything that helps you escape garbage companies run by garbage employees is good!
The same thing happened to Imgur where they slowly stripped features from the mobile site to force users onto the app, culminating in preventing mobile users from even logging in at all.
There's an option in your reddit settings to force them to serve you the old design rather than the redesign, which is remembered for future sessions. The old design isn't "mobile friendly", i.e. it requires a bunch of zooming in and out, but other than that it's completely functional, unlike the redesign which simply doesn't even work on mobile.
Reddit is now in a position where they pull a Digg and not suffer the consequences of Digg meaning they can be considered too big to fail. I guess this can allow them to pull whatever they want. Break api used by Android developer. Shove meaningless prompts on users face.
> then expect users to engage with it because it exists
Kinda like saying something must be done because it is "company policy"; passing the buck in the hope you don't question it, and realise company policy is also decided by the company.
> Users are able to identify when something is valuable to them.
I strongly disagree with this. Users are just responding to dopamine sensors that have triggers that they don't understand. "Value" extracted can be detrimental to a person's well-being, even if they chose it for themselves. Gambling addicts at casinos, obese people at fast food restaurants, pay-to-win whales and the newest version of Candy Crush. Reddit is just a casino where you gamble with your time and win addicting emotions like novelty and outrage.
That's not what "valuable" means in this argument. Reddit is huge because Reddit's diverse and novel content is valuable to people, as in, it's valued by them, that is what that phrase means. Doesn't matter if _you_ think it is _actually good for them_ or if it "produces value" economically. Yeah, it hits dopamine buttons, so what.
Using the same sense of value, Reddit's _app_ is not valuable to people relative to reddit-on-mobile which is why so many people in this thread agree with the claim.
> I don't understand why companies constantly do stuff that serves only _themselves_, and then expect users to engage with it because it exists.
Probably because most users don’t really care and those that do make excuses like “what else am I going to use?” or “I can’t stop using reddit, how will I stay in the loop?” or any one of many other excuses. If you want reddit to change, stop using them. Entirely.
I see a lot of HN users living in HN bubble completely forgetting that we are a minority of consumers. Most of Reddit audience is either not educated enough, don't have enough exposure to see the difference or simply don't care. Did we all forgot about eternal September already? So, unfortunately, no - users won't even notice this because they have no idea what's valuable to them and they will buy anything as long as it's packaged nicely. For the same reason they also won't vote with their wallets even when (as you say) "everyone" knows that testimonials are garbage.
> I see a lot of HN users living in HN bubble completely forgetting that we are a minority of consumers. Most of Reddit audience is either not educated enough ...
As someone who’s worked with the Marketing Dept. a lot in my prior role, the reason e-commerce sites (and possibly Reddit) do the things you mentioned is because they’re overly focused on new user/customer acquisition vs keeping current customers/users who know the the products’ value happy.
There’s almost always a bit of political drama behind the decision too: Marketing gets to say that they’re acquiring new users in short term (thereby increasing their budget and clout in the Org) while destroying customer value in the long run
What's equally absurd is that I assumed they ruined the web experience with their redesign in order to make it mobile compatible. It just goes to show they're ruining both web and mobile experience to pigeonhole users into environments where they can force feedback loops of their users so they have to spend more time in either platform.
Other people are saying "ads," but the story is probably more expansive. Mobile apps have better hooks for pulling people back into the app, like popup notifications. Reddit has natural lifecycle points where they can do this: when people respond to your comments or comment on your posts. They could also notify users of events like their yearly Secret Santa. I wouldn't be surprised if their engagement numbers were much better on mobile: since they can draw users back into the app with notifications, they have better ad views, more people buy Reddit Gold, more people comment (which leads to other people having content to read), etc.
This is correct. It's about retention more than monetization.
With the app, you can not only send notifications, but you also have the app icon which reminds people of the apps existence. Even little things like "badging" (when the app icon shows a dot or number to indicate new notifications) and even that can have a noticeable impact on retention (we're all programmed to click into anything with a notification).
Also, they can recommend better content for you. I don't know if this is still possible and whether reddit does it, but an app used to be able to get the list of other apps installed on the phone, and many companies used that as an input into their recommendation systems (along with your location, etc).
Also worth mentioning that there is no official API for Reddit's chat feature. You have no choice but to use the website or official app if you're on mobile and want or need to use their chat.
Also the chat feature is extremely watered down and doesn't even have support for attaching images. But that doesn't matter if someone on Reddit chooses to contact you via chat and you want to respond. It compels you to use their platforms instead of a third-party client.
If you always use third-party clients or old.reddit.com you'll never even know that someone tried to contact you via chat, which makes it somewhat ineffective at getting people to stop using those.
> With the app, you can not only send notifications, but you also have the app icon which reminds people of the apps existence.
This is exactly why I don't want to install the app. All of these things seem designed to steal my attention and direct it towards stuff others want me to focus on rather than what I want to focus on.
While having a mobile App store is probably a net good, in a lot of ways I wish Steve Jobs' initial vision of having the iPhone run WebApps and just improving the OS level support for improved hooks and functionality with actual websites was the way the world went.
The apps would never have gotten as good and stuff like Uber might never have happened. But it also would have kept Facebook and Reddit and all these other sites focused on making you visit because you want to visit rather than because you're being skinner boxed and nudged into visiting by overbearing tracking technologies.
I believe Apple is going to be allowing push notifications from the web soon, and you can also "Add to Homescreen" a webpage, and it acts like an app, so we might finally be seeing a needed shift away from the app store.
Apple added push notifications to safari, but only on desktop. That tells you that they deliberately do not support this because they know a lot of apps could be pwa’s otherwise.
Heard from where? Many people are clamoring for this, but other than wishful thinking I’ve never heard any compelling claims that Apple actually has it on their roadmap.
I, for one, would love to see support for push on mobile web on iOS. But I’m not holding my breath.
There's probably a suffering PM there whose bonus depends on juicing these horrible metrics rather than making their product wonderful. I myself dabbled in product management but got so turned off to it from the constant focus on addiction metrics. 7-day Retention! 30-day Retention! Daily Active Use! Daily Sessions Per DAU! Eyeball-share! Get their contacts! Spam them with E-mail! Spam them with notifications! Red badges on the app icon! Get them back in the app! I felt like a drug pusher, not someone building something great.
(Mentioned in another comment somewhere else in this thread). In the long-term, yes, but retention and monetization can diverge in the short-term. Classic example is showing fewer ads (better user experience, better retention... but much less monetization). Of course, in the long term, you'd still need to monetize that retention somehow anyway.
I don’t see how. Can you explain how a company makes revenue purely by retaining users? I think they still need a monetization scheme separately. [edited for clarity]
Very true and not actually pedantic. Retention is a pretty solid predictor of monetization, esp in the long-run. But in the short-run they can diverge. For example, decreasing ad units per page could drastically drop your monetization, but be great for longer-term retention.
Yep, I think ads is just another way of saying monetization. The app has lots of annoying notifications like "You might like...", "You got 100 upvotes", etc which, when you engage with more, they make more money through ads, coins etc.
Your attention, retention and engagement is a lot easier to manage and increase through an app than it is through website.
Instead of direct monetization, they probably measure by engagement. If a service has a permanent presence on your device, you get a LOT more engagement and recurring visits than a web version. Websites on mobile are one-time, visit-and-forget/move on things.
I'd love to be proven wrong though; what websites see daily, recurring, long time active usage?
Also not to be underestimated is that a mobile app will always have better UX than a website. Think of the subtle things like page transitions. Navigating Reddit still causes full-page reloads, whereas on the app it's a much more organic process. Speed is also an important factor.
Exactly this. To give you a datapoint from our properties: Our web site attracts 90% of the users, but only 60% of the actual traffic. In all metrics like retention, frequency and time on site, mobile leads by factors of 2-5.
Part of that is definitely self-selection, but as some people already said, other notable effects are push notifications, mindshare, loss aversion (you're on somebody's device already, so they can just as well use it) and partly better performance.
So much about the users, but Monetization is much better as well, mostly due to mobile ad-IDs, which especially on iOS lets you extract double the revenue per user due to targeting. Ad blockers are harder to bypass.
In the end, a mobile app will get you anywhere between 2-10x the revenue per user you attracted to your property, so that's why Reddit is pushing so hard.
I understand, but they're also pushing many users away with this behaviour.
It's what I like about HN.. It feels like they don't even care how much users 'engage'. I bet they don't even run a report on it. They just let us do our thing here and not worry about it. As a user this is a much nicer experience.
Easy for HN to do that. HN is not incentivized to make money. The people who run HN do not care if HN does not make any money. In fact, they have zero monetization strategies for HN.
HN has ads on its front page. Not many of them, and they are quite hard to spot at a glance (mixed in with the content, not clearly marked as such) but they're there.
AFAIK, the only ads are the ones described in the FAQ:
> A regular "Who Is Hiring?" thread appears on the first weekday of each month (or Jan 2). Most job ads are welcome there. Only an account called whoishiring is allowed to submit the thread itself. This prevents a race to post it first.
> Another kind of job ad is reserved for YC-funded startups. These appear on the front page, but are not stories: they have no vote arrows, points, or comments. They begin part-way down and fall steadily. Only one is on the front page at a time.
I'm not sure either of those qualify as ads in the sense we're talking about Reddit ads -- the first one doesn't bring in any revenue to HN (or cost "advertisers" any money); the second one seems like it's a perk for "graduating" YC and it's unclear whether it's something HN makes revenue from, either.
There's currently one on the frontpage for GiveCampus, shown below a regular story:
14. The Haunted House – Privacy on Google Street View (harpers.org)
8 points by b0b10101 1 hour ago | flag | hide | 1 comment
15. GiveCampus (YC S15) hiring Sr Engineers who care about education (lever.co)
1 hour ago | hide
Distinctive features: no upvote button, no point value, no flag link, no comments.
Right, I included the Haunted House entry (a regular story) for contrast.
The job posting is certainly "sponsored content", a privileged post put there at the behest of a third party, which blends in fairly well with neighboring posts yet isn't subject to the normal rules. I don't know if money actually changes hands to put it there (maybe each YC company is granted a posting there, or maybe they have to pay for it), but in other respects it's straightforwardly an ad for a job.
All that said, I have no problem with it being there.
Your answer reminds me; we are focusing on Hacker News as if it's its own thing, but part of the answer is right in the URL. HN is a part of ycombinator, which presumably pays the bills.
> what websites see daily, recurring, long time active usage?
Reddit
Do you really think that people would stop going to reddit if they used the site instead of the app on their mobile?
(But, of course, the people more prone to recurring visits are more likely to install an app.)
> mobile app will always have better UX than a website
And again, it's reddit we are talking about here. The app experience is just horrible.
They really hurt their own monetisation by making the subscriptions too expensive though. They more than doubled in price recently. It's not worth the price anymore for me.
They also keep adding new stupid in-app notifications (I have OS notifications blocked for them) which ironically you have to use the website to disable. "Trending" stuff in subreddits you're not even in and suggested subreddits and garbage like that. I assume that rubbish does become an OS notification if you let them.
Those kind of “nagifications” are among my biggest pet peeves on the modern internet. Almost every social platform these days abuses the notifications icon to show you a bunch of not-actually-a-notification garbage to boost your engagement. LinkedIn is especially egregious in this regard.
Their mobile website is terrible and intentionally crippled... I've found that when I'm having lots of meetings–whether it's sales, recruiting, or looking for work–being able to quickly pull up profiles of people I'm talking with while on the go is useful.
That said they badge the notifications icon in the header of the desktop web version as well–just now mine had 8 unread "Notifications", 7 of which were suggesting I congratulate acquaintances for work anniversaries or telling me about trending posts. Why can't that just be in the feed? Maybe because I don't actually ever scroll through the feed, because I don't care about the garbage people post on LinkedIn.
Yelp doesn't let you see more than a few photos in mobile web; if you switch to desktop view you can see all the photos but of course all the dimensions are janky.
This one in particular drove me up the wall when they started doing it, because, for at least a year, the Yelp app was so buggy (on my phone at least) that it was essentially unusable. I wanted to keep using Yelp on the web, but because it was so crippled I switched to using Google Maps as my default review-searching app. The Yelp app works on my phone now...but I never reach for it.
reddit notifications are useless now. I do use the app but all notifications are off. I only look at that in app notification thing if I recently posted and am expecting replies.
There are so many mobile apps out there that are little more than wrappers around the mobile site -- but it's vital that they be apps for precisely this reason. Neverthless, as a user, it drives me crazy; I just want to access the site in my browser.
What we are starting to see is web push notifications becoming fully supported by browsers, and I think it's only a matter of time before iOS and Android start allowing app-like websites to notify users without them having to download a full-fledged app.
It would be interesting to see what this does for the iOS and Android development landscape. They're in a golden age now where every major property needs both a mobile site and a dedicated app on both platforms. But if the mobile site can handle more features that were previously the sole provenance of native apps, that seems likely to shift the landscape.
Android support for browser notifications landed in Chrome mobile about five years ago. (And yes, that included support for Android versions that were old even then.)
iOS still has no support, and nothing even on the horizon, as far as I know. Apple is intentionally holding back the open web. I find this infuriating.
The constant nagging to "download the app" is worse. Notifications can be left off, for a website or an app - this constant install-app-nagging is unavoidable now.
It absolutely IS possible to add the feature, as opt-in like with app permissions, and e.g. with some user-initiated event like audio already does.
The REAL motivation here is money, forcing users to use the appstore, instead of the web which is an open protocol.
And reddits App-nag setting is not a permanent choice. When you disable it that's a localStorage thing only, so if you sign out, clear caches, sign in elsewhere, still nags you.
This is true. I usually remove all cookies and data from my devices every three of four days - it just seems tidy to do so, even though I usually open a private tab for following links. There is the slight annoyance of logging back in to a few web sites and turning off notifications. That is definitely an opportunity cost, since it takes a little time.
They support web notifications in safari in the desktop. Whatever annoyances would be caused on mobile are also present there. I am convinced the only reason we don’t have push notifications on mobile safari is because it would enable people to write pwa’s instead of apps.
I'm personally not a fan of notification spam either, but I would assume that notifications via mobile browsers would be opt-in. Is that a bad assumption?
I can't imagine user-hostile features like silently allowing websites to spam you would drive browser share in a wanted direction. That's like expecting people to use your website because you sell their information.
I don't want the trend of browser-as-operating-system to continue.
Aside from the shabby technological issues involved, the applications themselves these pseudo OS features are put to generally are opposed to some of the things I value highly: privacy and stability, for example. Most applications don't need a social component or to harass me to re-engage, to utilize hardware or other software services on my device, or even to export/share data with other applications.
As far as the technology itself: the platform is generally a terrible technology for even the dubious purposes it is usually put. It's clunky, and even with things like wasm the development story for the platform is, at best, terrible. Also, in my experience, the quality of development in the web sphere is generally (not always) poor compared to similarly complex projects in other areas/applications.
If an app is on the Web, it's my choice whether I use it or not. I don't need Apple's permission and I don't need to pay them for using it.
Also, Apple's stubborn insistence on lock-in (and lock-out) is really annoying if many of your contacts are not on Apple platforms.
There are clearly trade-offs, and there is a lot of valid criticism of the Web, but I'm really glad that the Web exists as a sort of operating system that guarantees some freedoms that native platforms are increasingly taking away from us (although not all of them are as paternalistic as Apple).
Why's that? At least on Android you're free to choose another browser that doesn't support notifications, if they bother you so much. Or you could just turn them of.
I appreciate a platform that doesn't add these kinds of misbehaviors in the first place. It means I have less to watch out for or worry about in the future.
Fortunately Apple can't prevent PWAs from telling users that Apple is the reason they can't have good things. I'm disappointed however in Apple also blocking PWA's from the App Store.
Despite what many might think, everything doesn’t need to be an “app,” PWA or otherwise. Reddit is literally a message board, one of the earliest types of website. It doesn’t need my location — ever. Email works just fine as a notification, if I want it as a “push” notification, I could VIP-enable the notification email address in Mail. Does Reddit need my accelerometer? Or direct access to my camera or microphone? Or address book? Or graphics APIs like Metal?
Nope. It’s just a message board. Sell contextual ads for the different subreddits, don’t bother tracking people, and be done with it. Of course that would likely require a significantly smaller workforce.
Well, good. Being spammed with notifications is the primary reason I refuse to install many apps. Sure, you can turn it off but they really make you hunt around for it.
Yes, that's one big reason I avoid installing apps. But the requests are also a pain. Can I permanently opt out of all requests across all sites, and never see a page ask again?
Reddit doesn’t care about being less intrusive. If they did, they could stop the aggressive app-nag today. A PWA doesn’t instantly make them want to be subtle.
it's more expansive, as in more data that can be mined for adverting purposes. I got a presentation from Reddit a few of years ago on their ad-platform and it was terrible (for enterprises). Their sales team were hyping big changes and things that were coming including geo-location, strong user profiling etc.
This was pre-redesign and it became clear what they were doing and their roadmap of features all were to grow the ad business. Mobile makes sense as it's how their key audiences consume content and it enables stronger profiling, richer engagement and better ad-targeting due to persistance.
The mobile app allows reddit to mine user behavior (every outbound link click is tracked, even on the desktop website) and then they're able to link your device profile, to content you consume and engage with. This can be used for ad-targeting or sold to other data brokers to build a richer profile about ABC user with this device, across XY IPs, typically based in ZZZ location.
A guy I know worked for a PC manufacturer years ago. The offered a windows version of a machine, and folks asked why they didn't offer a linux version.
A linux version would cost more money, because they made a non-trivial amount of money by loading up the PC with shovelware.
I've seen this countless times at companies. The decision makers do not want to greenlight projects that do not make money, or worse that cost them money.
So I think if you follow the money you'll find a mobile app makes them more money.
Yes! And as a general rule, it's safe to assume their marketing department isn't interested in that issue. Marketing departments can't help but market themselves too, and if BS numbers make them look good even to the detriment of the company, it's almost a sure bet you'll see them do that.
I’ve given up on reddit on mobile. From time to time I drop in on it using Safari on the desktop, but it’s not worth it to me to deal with their engagement-ification in their mobile app, or obvious crippleware of their mobile web experience.
I’m not pissed off about this: It’s like dropping in on a bar I used to enjoy a decade ago, to find out it has been redesigned a few times and everyone’s a new face.
Sometimes, that’s a really cool experience: “Say, this is neat.” Sometimes, not. But times must change, and we either change with them, or get left behind.
I am not going to change and become an engagement addict, clicking on shiny baubles out of boredom. So in reddit’s case, I am a part of its past, not its present, and I must accept being left behind in its wake.
> It’s like dropping in on a bar I used to enjoy a decade ago, to find out it has been redesigned a few times and everyone’s a new face.
That is honestly the best analogy I have heard to describe my feelings towards the site. It's still reddit... sort of, but its not my reddit. I joined during the digg exodus and I don't know if it's because I was younger or what, but for a while reddit seemed like that go-to bar: always warm, inviting, and socially enriching.
And today, the once best feature for finding new stuff: the front page / popular, is just misery porn and American politics. If you don't already have a healthy list of subscribed subs with actually good content, I find that it's really difficult to find new stuff.
I've been using Sync (formerly Reddit Sync) on Android for quite a few years and love it. I paid for the Pro version (only US$3.99 one time payment) a while back. I turn off almost all notifications (do this for like 90% of apps anyways) and I only use it when I want to, not when I'm being asked to. It's very customizable.
I'm just waiting for the day that Reddit decides to lock down their API enough to hurt all third party apps and force people onto theirs.
Me too. I used to frequent Reddit occasionally on mobile web. I didn't, and still don't want the app for similar reasons, but also so I wouldn't get addicted. These days I just don't visit it at all because the friction is too high.
I've used Reddit Is Fun on Android for 8 years. Still looks the same and works just as well as it did then. None of this new design crap. Never found anything better
I really like Joey on mobile (Pixel here, not sure about iOS). No pings and is completely silent until I feel the need for wasting some time. It's shareware so please think about contributing if you agree.
Web browsing these days really feels like wandering through a market filled with panhandlers and pickpockets—you're just trying to look around, and everyone's up in your face demanding stuff in the most obnoxious way possible.
As a web designer/developer, this approach offends me. At the risk of sounding like I'm taking this more seriously than I should, I believe that good design is founded on respecting ones' users. Reddit throwing what's essentially an ad in my face every time I try to use their site is bad design because it is disrespectful to me, the user.
Maybe their app is a better experience; there are more respectful ways of highlighting it. As immature as this sounds, not using the app is almost a matter of principle for me at this point, because I don't believe in rewarding disrespectful design.
This is extremely interesting. Could the webbrowser of the future be the one which sways the nagging boxes away, and lets you access actual content? How would we reach that? (or, as more probable: this is the end of the webbrowser, it will die of a million cuts, and all smart TVs are on specific apps, all mobiles run on apps, and the one simple-to-use website that will remain forever will be... HackerNews).
Edit: And the only way to sway those nagging boxes away is to have a code of conduct, which is how appstores can impose a no-nagging experience.
I use firefox on android and immediately switch to reader view. It works wonderfully. The content takes up the entire screen, text always stays where you want it, scrolling response is instant and my phone stays cool while reading it. Plus you have dark and light themes for browsing.
Unfortunately, Firefox for iOS is built on Webkit, so there are no plugins for it, so this approach doesn't work on my iPad.
By now a browser is just a UI platform for a website, so if the website doesn't give you the content, the browser can't extort it from the server.
So far we can filter out some ads because for sites it's just cheaper to use off-the-shelf solutions from ad networks, which are the same for everyone, and to preserve accessibility via HTML. Enter heavy randomization or the canvas, and filters won't work. Look at Yandex: it took the fight with adblockers seriously, so uBlock's lists finally gave up—it just was no use.
If you try to use customization via CSS, on some modern sites the crazy mess of divs with cryptic classes already makes no sense and doesn't lend itself to CSS selectors.
You know, I could see—if this isn't out there already—a browser based on eliminating visual annoyances; essentially, something that operates a lot like iOS reader view, maybe like Brave but for popups and similar interruptions.
My current anti-crap shield is a patchwork of various plugins/utilities, most of which only work on my desktop: a plugin to turn JavaScript off (this alone eliminates about 90% of the nonsense, while preserving most of the valuable content); a bookmarklet that hides fixed-position elements (wipes out a lot of the popups/reminders); and a plugin that lets me edit a site's CSS so I can manually remove the junk that the first two methods don't. (Using adblockers and readability plugins goes without saying.) A browser with all of those baked into it—especially, say, something like smart popup removal—would be something I'd be willing to pay for.
I don't believe the web is headed for death; not yet, anyway. There are certainly trends that are deeply troubling, but for most of us, building a website is still far more feasible than building an app.
Use an RSS reader. You will like the lack of feeling of being forced to comment on things you don't care much about in retrospection. No tracking or ads, only content. You can filter shady sites or posts from appearing.
I built a better rss feed for reddit to use with Feedly.
Notable features:
* Links to the content rather than the comments
* Embeds a summary of the content
* Supports images, gifs and videos
* Extra query params (nsfw block, up vote limits)
* Open source https://github.com/trashhalo/reddit-rss
Yup. RSS is the way to go for browsing Reddit without all the dark patterns.
Self plug - you can try my app Plenary on Android (no ads/trackers) - https://play.google.com/store/apps/details?id=com.spians.ple... that has Reddit as one of the RSS Assistant option. It basically creates RSS feed for subreddits, users, search terms etc for you.
> Interestingly parent’s plug is a dark pattern to avoid dark patterns!
Sorry I didn't find any dark patterns here. If you can elaborate more on this, that would be really helpful.
> It’s also surprising that even on HN, a place where many people write software for a living, people are avoidant of advertising paid features.
Exactly! A lot of users want privacy focused - no ads apps. But many of them are hesitant to pay for it. For them, all apps should be FOSS (that means relying on good will of the developer that he will have motivation to maintain the app regularly or relying on some bug foundations) and no dev should generate revenue from their apps.
> Sorry I didn't find any dark patterns here. If you can elaborate more on this, that would be really helpful.
I don't think you were being malicious, but it would of been useful to know up front that those RSS features cost money is all. Otherwise I see nothing wrong in your initial comment.
You're right, I do want a FOSS solution and I'm hesitant to pay for something closed and controlled entirely by the dev.
But I am not opposed at all to pay for apps that include the source (and build instructions). I don't mind if it's licensed accordingly to prevent me redistributing it (which would undermine sales). I frequently buy apps like that.
More times than I can count I have bought an app that was later either abandoned when it stopped making money, or got an unwanted UI overhaul that wrecked it, or removed an obscure feature I depended on, etc. I will happily pay for software, but if it's not software I could potentially maintain myself then the amount of money I'm willing to spend goes way down and I'm a lot more hesitant to make the initial purchase.
> that means relying on good will of the developer that he will have motivation to maintain the app regularly or relying on some bug foundations
As opposed to paying for it which guarantees you'll get the service you want - except it doesn't.
At least FLOSS would give you and others the opportunity to fix bugs /maintain the app themselves.
> But many of them are hesitant to pay for it
Depends what it is. There's a difference paying for a non free app, and tipping for a free app.
Thank you for trying it out. Plenary has monthly, quarterly and yearly subscription options and even has a lifetime purchase plan. That is the way support the development of the app so that it doesn't have to rely on ads and works without any intrusions to users. That is the point of this post, right?
This. If you want something already hosted try Innoreader (https://www.inoreader.com). I like having software I can just support with money instead of seeing garbage ads and having my personal information sold because it's "free" (though they do have a free plan if you can't afford or don't want to support them).
Thanks for those links. Seems there have been some developments since I last checked into rss clients. Are you aware of any good browser plug-ins in that field? I've been using brief:
this is not my project but I have been enjoying this rss proxy which can let you limit which posts get into your feed by top rank, score threshold, or quantity of posts per day. I use it because even a moderately popular subreddit can be a bit of a firehose.
As a conversion optimization question, I suspect Reddit would get better conversion on their "do you want the mobile app" banner if they only showed it randomly about 10-15% of the time. By showing it every time, they cause viewers to remember their "why would I want that - I don't" decision from the previous time they saw the question. Reducing the frequency would likely result in more people considering the question as a fresh new question independent of the previous viewing when presented with it the second time, increasing the odds they go with the install (by definition these are people who didn't install the first time, so their remembered decision is by definition no, so encouraging them to consider the question again as a fresh question can only improve the conversion effectiveness of the prompt over their remembered decision).
I learned from working in marketing is you have to let go of your biases and assumptions about user behavior. In the beginning I found myself constantly thinking "why would anyone in their right mind click on this ad", but we know people click on ads. Most of us in tech are very different from your typical internet user at large.
We have to assume that smart people are working at Reddit, that they have tested the living crap out of this model, and have found it to be optimal.
Please don't suggest them more ideas to make this even more annoying than it currently is :) (and yes, I would consider reducing the frequency of displaying it as an increase in annoyance, for exactly the reasons in the parent comment)
How are fewer interruptions more annoying than more interruptions? I’ve reread the parent comment a few times and I don’t see how it could be worse than it is now.
I believe the parent commenter's logic is that reducing the frequency of the presentation increases the amount of decision making the viewer goes through when it is presented. The commenter's belief is that the added decision making effort would be more annoying than the benefit from the reduction in frequency.
Anti-tracking cookie defenses will mean that, for sites you don't visit with sufficient frequency, any remembered decision will be forgotten a week after your last visit.
I believe this is why cookie warnings have suddenly skyrocketed, and why Google Recaptcha treats us all as robots suddenly. It means that the form of 'intermittent' tracking you describe will only work for everyday Reddit visitors, not for those sufficiently occasional to see their cookies purged.
Reddit shows the same fucking message on nearly every page, which is the worst possible implementation you could have. As with others commenting here, I refuse to use Reddit on mobile anymore.
You don't need to track users forever on end to only show messages a small percentage of the time. Just showing it sometimes on a new visit is good enough to not be annoying. Even better if you wait for a few site interactions before even considering displaying the message. Also, don't show messages more than once for users who have logged in.
This is not a difficult problem to solve. Companies just refuse to understand that no-means-no.
What you describe has been the case for years if you use "old reddit" on a mobile device. There's a big "GET REDDIT MOBILE" banner at the top of every page that disappears when dismissed, only to reappear randomly a few days or weeks later.
I have nothing against you, but this is always super funny when you get some random people suggesting how to improve the conversation rate of an app, which probably have several people working full time on the issue.
Yes there are always ways to improve, but some people working there are expert on this specific topic, if the topic is important for the company they probably already do what is the best
Did you know that companies that get on top 10 of most stock exchanges around the world don't actually stay there in like 10 years? They slip drastically, anywhere in the world.
Yeah, so companies aren't led by super humans. They make stupid decisions all the time.
That's probably because those decisions are hard to get right, not because the leaders are stupid. And even harder for a random opinion-haver than the CEO.
> if the topic is important for the company they probably already do what is the best
Having worked at many companies, I find that statement very bizarrely naive. I don't understand how anyone with any real world experience could possibly actually believe what you said.
I'm never going to install it, and I have all but stopped reading it because these prompts are so obnoxious. That's probably +$ for Reddit though.
I don't understand why companies constantly do stuff that serves only _themselves_, and then expect users to engage with it because it exists. Users are able to identify when something is valuable to them. If you make it valuable they will use it. Consider the difference on an ecommerce site between a comment section vs a few company-picked "testimonials" above the fold. _Everyone_ knows the testimonials are garbage. Maybe your conversion goes up a tick the first time you put them on the site, but when a repeat viewer sees the same ones again they're going to roll their eyes and register you as untrustworthy. Whereas a (reasonably-managed, honest) comment section provides loads of information that's actually valuable to the consumer.