I believe this is the point, rather than being mundane. Other video conference tools are not able to offer this debugging option - which you have pointed out is useful.
I was part of this test - I've refused to install the app for years (I prefer websites and the option to open tabs for later browsing).
It was extremely annoying for a site I've used for 12+ years to treat me like that. It did massively cut my usage of Reddit (which I would consider to be quite high), I primarily access it from my phone and I all but stopped using it for the week or so I was in the "experiment".
> I prefer websites and the option to open tabs for later browsing
The number of times I lost a post because when I switched app or didn’t use my phone for a while reddit would just reset to the home page… I think they really didn’t realize how much the shitty UX would enrage people against them.
Would be faster to just shut down the corporation and give the money back to the shareholders, whatever they have anyway, but your approach is not without merit, mostly in that I think it would be funny at this point.
That's basically why I'm advocating what I posted, that and also because I think it'd be a great chapter in tech history, showing how doubling down on a stupid user-hostile decision can destroy a company (assuming it does). Then, 30 years from now, people will still talk about how Reddit self-destructed when they decided they needed to control the user experience and forced everyone to use their app.
Rationally, your approach is of course better, but the current narcissistic execs aren't going to do anything like that, whereas I can certainly see them doing something as stupid and out-of-touch as forcing everyone to use their app on both mobile and desktop.
> Then, 30 years from now, people will still talk about how Reddit self-destructed when they decided they needed to control the user experience and forced everyone to use their app.
No. What they'll be talking about is how Reddit had too many holes in its fence, and all its cows escaped. They'll then discuss all the advancement in cattle fencing and barn construction that happened in the 30 years since.
Why do people assume Reddit C-suite and investors are being stupid or narcissistic here? That would make sense if the relationship between them and the users was a friendly one. It isn't. It's adversarial. For Reddit (as well as Meta and other social media platforms), the users are cattle. Even worse than that - they're stochastic cattle. Nobody at the top cares if you or me are having a nice experience with the site/app. They care about the value extracted from us in aggregate. To them, it's an optimization problem, and it's been apparent for a long time now that the optimum point is usually "the most shitty and abusive possible version that still clears the 'fit for purpose' bar" (the end point is more obvious when you look at goods and services that have been around for a couple decades or more, and thus subject to decades of "value engineering").
It doesn't feel as bad when they're optimizing for future value extraction, but that time is past, and Reddit is currently squeezing value out of its cattle-base.
Is it sustainable? Since when did that question mattered to the captains of the industry? "Reddit" as a brand and company matters to the users; for its leaders and investors, it's just a money making instrument that takes time to mature, but exists to be squeezed, discarded and replaced by something else.
That's exactly my thinking. They can make a special Reddit app with Electron that basically recreates the website, but forces users to see lots of annoying unblockable ads while hogging lots of memory. Users trying to use the normal website will just be directed to download and install this app, and only shown a preview of the site that they can't use. What could possibly go wrong?
For all this UX design, responsiveness, mobile-friendly talk and SAAS products for conversion funnels and what not, does no one in management ever use their own app or website on mobile? Seriously, open practically any random website on a stock chrome or Firefox on mobile and just see how horrible it is. Scrolling loses position, you randomly click stuff and shit happens, half the page is filled with a sign-up newsletter popup or a privacy banner. And wow if you have to input anything and all the nonsense you have to put up with dealing with the page or app responding to the keyboard. Or you have to scroll in an input field.
Like seriously. What is wrong with these people that designed this shit and how do we not have an alternative.
Or clicking on some post. Nothing happening. Trying to click again, but new screen appears from original click, so your second attempt lands on some new link.
The whole auto-refresh thing is bullshit in general. You see something interesting, and then two microseconds later it is gone because the app (or even tab) decided to refresh and bring you new content.
That's the point actually. And not only in Reddit, but in all current social networks. You must lose posts, you must always look at the fresh snapshot of posts. And it works, people are conditioned to work with ephemeral internet. Apps and sites all work like this nowadays - Facebook, Instagram, Shitter, Reddit main page, Netflix main page, even HN partially. You blink and everything is gone, here is new content for you, enjoy, but not too long and don't become attached.
This drives up engagement in the population with attention disorders and promotes advertisement, since it is organically natural to see ads between ad-like endless posts.
Yelp started doing this years ago. They slowly cut access to their mobile site that used to work perfectly well. Sure I could install the app, but the obvious disrespect to the user pretending that the mobile site didn’t work perfectly well and that you have to install the app was so frustrating that I just stopped using Yelp altogether. And that’s saying something, seeing as I own a local theater venue and refusing to engage with Yelp hurts be more than it hurts them. But screw it, I hate being bullied by these platforms.
I was on reddit before they tried updating their designs, the only reason I'm still there is because they still have the old.reddit.com frontend available. I even use it on mobile where it's not exactly practical. It's not because I have some sort of aversion to change, well, I guess I'm really uninterested in downloading apps considering I didn't even bother to try things like Apolle to see what the fuzz was about, but their various attempts at redesigns have been so bad that I would rather use old.reddit.com than them on mobile, even though it's impractical.
On a computer I see no benefits from any of the redesigns compared to old.reddit.com. I work a lot with Typescript and also React myself, and I love the language, so it's not because I dislike that sort of thing, but I think a list of links with comments just works better without being put into a virtual DOM or even just JS. HN is the perfect example of that, there has been a lot of hobby JS frontends from people, but they all work worse than the real deal and somewhat hilariously they work better than reddit's professional attempts. Now I get why reddit wants to move away from the page-reload. They want a lot of the SoMe interactivity, like their silly chat and so on, but I'm not sure who would ever want a Facebook with total strangers instead of people you actually talk with. I sure don't.
Could someone explain why a new web interface, albeit arguably better for mobile and actually enjoyable on desktop if the following can be forgotten, is so damn slow? When loading it appears to emerge from unknown depths and open up with a heavy sigh. Personification of a tool but this the impression it gives me every time. I thought 2020s were years when multi cores and gigabytes of memory would render everything snappy.
It's built by the idiots sons and daughters of rich people as their "my first job" project. Lots of sectors apparently function like this. Children of the elite can have play jobs, money is distributed to friends and family and everyone suffers.
There were posts on reddit about the most toxic work culture there i remember with drugs and bizarre politics straight out of some san fran sitcom.
And some quite funny posts on just how grotesquely "my first react project" the whole code base was, and still is. These people are absolutely clueless. They nuked the whole "new reddit feedback" forum with posts pointing out just how bad the whole thing was, like pulling tens megabytes of starter boilerplate in production and loops with script loads inside of them that could grind a powerful computer to a halt.
I signed up for Reddit last year, finally. I’m a happy user of old.reddit.com. And if it goes, I go. I survived a decade without an account, and it would be easy enough to go back that way. My opinion is not that important to share.
Old Reddit with a Stylus theme is how I've been using it for years. I'd occasionally switch to the new site just to check it's progress and while it has gotten a bit better in the last year or so, little nitpicks eventually drive me back to old.
As much as old reddit is a clear winner ok desktop it is pretty awful on mobile. Personally I actually prefer the new site on mobile (although it is awful too) but I understand why sone people still prefer the old site on mobile.
People refusing to use their app etc. are likely a minority. Most people are completely used to being bullied into submission by tech companies, and will happily follow along.
I'd assume their goal with this isn't to convert the stragglers, it's to just close the gates to them so that they disappear from ad-view related statistics.
Edit: Further to this point, the Apollo app which everyone was talking about the other day has 50000 (fifty thousand!) paid users. Reddit has hundreds of millions of monthly users. They don't care about this minority of users, they just want the shitstorm to pass so they can move on. They also don't care that there's likely a small minority of users creating most actual good content, but it doesn't matter because the site can be floated entirely by meme spam bots and porn posts and still be massively profitable.
I'm personally not confident if the signal to noise ratio took a turn for the worse (re: spam and porn posts), the site would be as a profitable/popular anymore
What I'm curious about is why didn't (and maybe they did as I'm not particularly well read in this situation) reddit just buy the apollo app for like $1M/$5M/etc. and then just modify it so it injects whatever tracking they want instead of creating this giant far more costly shitstorm?
Further, maybe the 50k paid users but perhaps there were many unpaid users using it less intensely?
Fortunately their app is compatible with the ReVanced project (think YouTube Vanced). It's terrible you have to recompile apk's to get a useful experience (sans ads, sharing tracking, other restrictions), but for me it's currently the only viable way to use Reddit.
Revanced is the only way to stay sane while using Youtube and Twitch on mobile. The adverts and other spam are more obnoxious than cable TV.
I just hope it flies under the radar enough that Google don't start banning accounts for using it. I don't log in to the Youtube app for this reason, but I'm sure there is a line in a EULA somewhere that means they could if they wanted to.
Which reminds me, i need to backup all my email...
I stopped using reddit when the .compact interface was removed. No I go back once in a while but if this goes through I will probably simply never visit.
You can still use old.reddit.com, which is vastly superior and faster than the modern reddit. It also loads videos and images inline (when you click for them to load) and it de-emphasizes comments. This allows you to work like the compact feed where you scroll through posts and only go into comments sections if you _really_ want to. It is a good alternative.
It's funny because I also sing the praises of old.reddit.com but would describe it/my preferences in a totally different manner. My preference would be for reddit to be as close to HN in UI format as possible, perhaps with some minor thumbnails for visual content.
What I like about the old UI is that I find it emphasizes the post titles instead of content. I *don't* feel inundated with images & videos, besides the little thumbnail. I usually just want to skim titles, not look at visual content, and I find that impossible in the new UI. And I find that it's easier to get at/see the comments than in the new UI.
What happens if/when another country (China? India?) introduces a similar law, requiring a company that operates there to make data on u.s citizens available even if its stored outside of the country. Will these companies stop operating in these markets?
Buzz was actually a nice bare-minimum functionality of social networking - lite Twitter. Plus is Google trying to consolidate some monstrosity of everythign Google Social into one space.
Buzz was fine, if ugly. The only real problem with Buzz was how they leaked private GMail information.
I can appreciate what wordpress has achieved over the years, and what it has helped to create. But the current wordpress code base is a nightmare to work with, and doesn't scale well.
Having said that, it is an excellent blogging platform and CMS, but this is something that is usually forgotten, resulting in it finding itself shoehorned into the most inappropriate places.
I have scaled Wordpress to multiple front end servers with not to much work (because Wordpress is stateless and doesn't have sessions you dont even have to worry bout sharing sessions).
The only thing you need to worry about is having a shared storage for the uploads. You can do this through a NAS or I just upload to s3 then use a CDN.
If you get to the size that you need to have multiple databases (which you shouldn't if your using a page cache plugin like the official Batcache plugin) wordpress offers HyperDB as a solution.
I am actually going to be working on a Wordpress-a-a-Service type hosting solution where speed and scalability will never be a problem for the customer.
Where Wordpress does not scale is the default install on shared hosting or a VPS with little resources.
Finally I just wanted to thank Matt and the whole team at automatic for all their hard work into a maybe in-prefect but much used work-horse of the internet.
> Where Wordpress does not scale is the default install on shared hosting or a VPS with little resources.
True for shared environment but WP does fine even on a low-end vps. 2-3 years ago I helped someone setup wp on a lowend vps server which successfully handled ~5million pageviews under 24 hours, the vps had a measly 1gb memory and the memory usage never went over 700mb. nginx/php-fpm/varnish/apc/w3t done. Took me less than 1 hour to set it up.
I don't understand when people complain about wordpress being bloat but at the same time wants it to solve all kinds of problem right out of the box.
Unfortunately most of the issues come from WordPress being written for a web that existed 10 years ago. Static pages with little dynamic content.
This doesn't really work on a more "modern" style of site, where the content is more dynamic and changeable.
Again, a lot of these issues only really arise when your product moves past being a blogging-style platform (such as happened to us), but this is something that either WP, or the WP community, seems to be striving for more and more.
Yes, yes, the famous hosting environment argument. It comes up whenever the WP team just don't want to do work on the actual guts of the system as opposed to tickbox features.
When they added autoupdate, that had large host environment implications. Import/export has host environment implications. It goes on and on.
There's no technical reason they can't have a simple page cache in mainline that turns itself off when there's no write access.
Not to mention that WordPress is now popular enough to bully hosting companies if they wished to do so. What hosting company would want to admit that their services are incompatible with the latest version of WordPress? WordPress is no longer under any realistic obligation to respect the lowest common denominator of hosting environments, because whatever it requires will become the norm. It's a trememdous power that few open-source projects enjoy, and one that could be used for the greater good. If the next version of WordPress just went ahead and required PHP 5.4, for example, the hosting industry would have no choice but to upgrade their PHP versions a.s.a.p.
A lot of software developers are seduced by the old "80/20" rule. It seems to make a lot of sense: 80% of the people use 20% of the features. So you convince yourself that you only need to implement 20% of the features, and you can still sell 80% as many copies.
Unfortunately, it's never the same 20%. Everybody uses a different set of features. In the last 10 years I have probably heard of dozens of companies who, determined not to learn from each other, tried to release "lite" word processors that only implement 20% of the features. This story is as old as the PC. Most of the time, what happens is that they give their program to a journalist to review, and the journalist reviews it by writing their review using the new word processor, and then the journalist tries to find the "word count" feature which they need because most journalists have precise word count requirements, and it's not there, because it's in the "80% that nobody uses," and the journalist ends up writing a story that attempts to claim simultaneously that lite programs are good, bloat is bad, and I can't use this damn thing 'cause it won't count my words. If I had a dollar for every time this has happened I would be very happy.
When you start marketing your "lite" product, and you tell people, "hey, it's lite, only 1MB," they tend to be very happy, then they ask you if it has their crucial feature, and it doesn't, so they don't buy your product.
Wordpress seems to have everyone's critical feature, or critical plug-in, or a developer intimately familiar with the platform who can be hired to write your critical plug-in. By the time "more specialised blogging engines" have the flexibility and pervasiveness of Wordpress, I bet they'll resemble. . . Wordpress.
I'm not fan of the codebase, but until something else comes along that's as easy for my clients to learn/use - I'm going to keep on recommending and installing WordPress for them.
Anybody who's used a word processor can pick up most of what they need to know to work as an author or editor in WordPress in a afternoon. I'm not about to recreate all the widely available documentation and tutorials WordPress has available - or explain to my clients "all you need to do is write all your website content in MarkDown, then run this Ruby script from the command line to publish it to S3/CloudFront!"
I imagine the reason most people use WordPress isn't because of the authoring or editing — most of which can be replicated easily using things like TinyMCE — but the ecosystem of themes and plugins. Your technologically-inept client has access to a vast number of free, cheap, and easily installable themes to style their website whatever gaudy way they want.
There are plenty of better publishing solutions in terms of codebase, and there are plenty of solutions that offer equal or better user interfaces, but none that bring the ability to choose from hundreds of thousands of themes to style a website, which is the main focus of your average blog publisher.
If other specialized blogging engines had one-tenth the extensibility that WP has, then I could see myself switching. But when you have hundreds of thousands of people willing to write plugins for free or very little money that integrate with the most complex payment systems and third-party APIs, you tend to deal with the crappy core code of the main module.
Those banners (in most cases) are not you turning on or off cookies. The bbc sets cookies immediately (they do allow you to unset some here http://www.bbc.co.uk/privacy/cookies/managing/cookie-setting...), but you already have a session cookie that will last 4 years once the page is loaded.
Most sites will state they are assuming consent based on your continued usage of the site.
How do you do that? Once she joins the team, you have no way of knowing if something Ann created or modified belongs to the team or to her. This is the classic problem with "taint checking." Once you mix data across boundaries you have to assume it's tainted. I agree it's a crappy situation, but I can't see what Dropbox should change, other than make it very clear how this feature works to future customers. As a one-time thing, they could maybe get a PR win by helping this guy recover his data (if possible). However, that's not scalable, or a precedent they'd want to continue forward.
They can clearly delineate between team and personal accounts, force team storage to certain directories, and/or not allow users to manage team/personal accounts with the same login.
I was specifically responding to the previous comment, to explain why Dropbox can't simply revoke access to team storage files at this point (because they can't guarantee what constitutes team data). However, I'm in agreement with you that Dropbox could introduce new features to prevent this situation from happening in the first place, and had already noted as much in a different comment.
Dropbox doesn't delete the files, so none of this matters. They're only deleting the account, which revokes nothing but causes lots of problems. There is nothing hard to fix. Just switch it back to personal quota.
Looks great, quick question someone may be able to answer: in the license he says "[free to use in] any personal or commercial work...Do not redistribute or sell", which is fair enough, but does that restrict use in paid-for apps (such as on mobile devices etc)?
The license terms mean that the icon set cannot be used in open source projects intended for redistribution. Your call, of course, just pointing it out.
It might be better to use a more standard license, maybe Apache or BSD with an extra clause prohibiting "sale in whole apart from use in a larger software distribution." IANAL but it might hold up better and be less confusing.
Otherwise, obviously really great work and very generous. Thanks.
@davidw - that is meant to refer to not selling or redistributing the icon pack as a whole - you are more than welcome to use it in any open source redistributed work.
I think that requirement would conflict with some open source licenses -- they typically grant the right to redistribute and modify the source code in any way.
That would include stripping out everything but the icons and selling them.
So if the license requires all accompanying code/assets to be under similar licenses, that could prevent use of your icons. I think that's the case under the GPL? Not really sure anymore.
e: After doing a bit more reading, I believe I was wrong. As long as the icons aren't somehow compiled into the binary, there are no license problems.
No, this differs from CC-NC in that you can use them in a commercial work, you just can't sell them on their own. Using CC-NC would prevent users from bundling these into a commercial product whatsoever.
Public Domain and CCZero would be a very efficient way to quiet people's fears about what they can't do with this icon set, without alarming them that they need Legal Advice to build their app. You've tried to trim down the restrictions to almost nothing, but we have an established protocol to encompass all uses. Why not go all the way?
