Yes, I use Kill Sticky and also this bookmarklet to clean up a lot of the cruft I come across. It's not a perfect system but it works more than it doesn't.
What're the benefits of removing iFrames over using an adblocker like uBlock? Seems like global iFrame genocide would hurt your user experience more. They're not all ads!
This is a bookmarklet, not an extension, so you only activate it when you need it. To answer your actual question: autoplay videos that follow you when you scroll >:|
Kind of random comment, but thank you! I switched to Firefox from Chrome a few weeks ago, and one of my major annoyances was that Firefox would play Youtube videos opened in background tabs, even before the tab ever got focus. Seeing your comment gave me the idea to search about:config for a setting related to autoplay, and I found a setting that enables my preferred behavior.
I work at a bank and not on the tech side so I'm restricted to IE 11 with no addons. iFrame genocide is as close to uBlock as I can get while working. At home of course it's ghostery and uBlock.
I find most implementations of this jumpy and distracting - just too eager. Often I'm scrolling up to reread the top line of text. Then the bar pops up and blocks my text, it is daft!
Awesome, thanks! Trying it; I've been thinking of building something similar for a few weeks. Site-specific (like the Medium addon we're discussing here, or domain-based userContent.css) solutions suck, I want something generic (like the bookmarklet) and automatic (thus the "tradeoff" of popping the dickbars on under certain conditions, like you did when scrolling up).
Ever thought of packaging this as a Firefox/Chrome (WebExtensions) addon? Want to collaborate on one? Or if you don't care, are you okay for me to start one from your userscript?
Heh this would just completely break one of my current projects (not using `fixed` for headers I swear). It's interesting to think about people out there using things like this.
Hacker news is easier to read on mobile than many so called made for mobile sites. It's as if made for mobile means: "abuse the small mobile screen to show as much as possible spammy bars and social network buttons on it and make them even bigger when you zoom in, while hiding actually relevant information that you can see on the desktop version of the site"
While IMHO mobile optimized should mean: wrap the text automatically at the width of the screen, in a readable size in portrait mode, don't hide anything that's visible on the desktop site (like dates, user comments, version history, ...), and support zooming in in a sane way
Even Wikipedia fails at showing the user comments pages and sane zooming in of pictures (it's always those sticky bars that ruin it)
Most "mobile" versions of web sites are objectively worse than their "desktop" versions.
Actual advantages of mobile devices are:
* High DPI, handheld. You can look at a mobile screen from different distances for different use cases and screen contents.
* Less eye movement required than for using a typical HD desktop monitor. This makes it easier to navigate unknown applications / web sites.
* Easier to tap anywhere than to click anywhere with a mouse.
* 360° scrolling with high accuracy and intuitive control over the velocity.
* Zooming with quick and intuitive gestures.
Mobile devices are perfect for navigating wide, complex web sites. That's why "mobile" versions that ignore the advantages of mobile devices are usually worse than the desktop versions.
How "mobile-optimized" apps / web sites often fail:
* Overly large elements and fonts that cannot be zoomed out from. High DPI goes to waste.
* Fixed elements that obscure parts of the screen. They make the scrollable area so tiny that you have to move your eye balls quickly when scrolling.
* Hiding important elements in hamburger menus or making them inaccessible. Finding stuff has to require multiple taps so that it is even slower than with a mouse.
* Horizontal scrolling is evil except in awfully implemented code snippets. It is unimaginable to the modern web developer that someone would double-tap a column of text to read it and scroll around afterwards to discover what else the site has to offer.
* Vertical scrolling is too hard even though a touchscreen is perfect for that. All the "important" shit has to be fixed because users would be too stupid to find the menu by scrolling. A fixed "back to top" button is the icing on the modern web cake.
* Zooming is too hard so it should be better disabled, bugged or useless.
I don't use my phone that much because the usability is rubbish, and when I do I find it an exercise in frustration.
So, when someone asks me to make a website usable on mobile, I really have no idea what usably on mobile would actually involve. And the result is I make a single column of content which is 100% of screen width and has all the buttons and text inputs enlarged.
I've tried looking for advice about this on the web, but because mobile is such a hype-driven area, the search engine results for any mobile web design topic are inevitably marketing idiocy.
So, where can I read something by people who've done real usability experiments on touchscreens?
> * Easier to tap anywhere than to click anywhere with a mouse.
> * 360° scrolling with high accuracy and intuitive control over the velocity.
> * Zooming with quick and intuitive gestures.
I disagree about those. "taping" lack a lot of precision a mouse has. And don't get me started on the typing experience as you can't use all your fingers and have no tactile return. Also Ctrl-F "the thing I'm looking for" anyone ?
The intuitive gestures are all but intuitive if you have never used them before. My first reflex to zoom is sure not to move diagonally 2 fingers on a picture.
I always forget that mobile browsers say "find in page" rather than "search" when you're trying to search for something on the page, so my eyes skim past it all the time in the hamburger menu.
On iOS you can put find in page in the bottom row of the screen that pops up when you hit the share button. I also put request desktop site there for easy access (although I recently accidentally discovered that that's available by long pressing the reload button.)
Long pressing the reload button ... wow ... yep it does that. I wonder: Is there some site that just lists all the different tricks for mobile safari in a list?
I feel like I am 8 again bombing random walls / burning random trees in Zelda looking for a super secret fabled (atleast in my grade school) green potion of unlimited refills when I'm using iOS safari half the time.
Also, I feel like double tapping the address bar used to scroll to the top of the page too but it seems like it doesn't do that anymore so showing what was ninja patched out would be useful too.
>wrap the text automatically at the width of the screen
This should be the browser's job, and none of the webpage's business. Opera Mobile used to do a great job at this, together with zooming and generally making sure traditional webpages rendered in a usable manner on a small screen. With the spreading of dumbed down mobile sites, overcomplicated layout systems and overall inflexibility due to webdevs being UX control freaks, we apparently decided that was not the way to go and instead settled for whatever inferior experience the individual webdev comes up with for their website.
Yeah, whoever, the whole ecosystem, I do have the "text reflow" plugin for mobile firefox which helps a lot, it requires clicking on every individual paragraph to wrap it through so it could be better.
Firefox + uBlock Origin on Android goes a long way toward mitigating the worst horrors of mobile sites, but sometimes even that's not enough. I'd consider setting up NoScript as well, but that would be a hassle to use on mobile. It certainly is a lifesaver on the desktop, though.
Except for verbatim/code blocks (i.e. line starts with two or more spaces) with, which are basically unreadable on mobile once they
stretch to multiple lines; I get that the intention is for code blocks, but they seem to be misused for quoting OP somewhat frequently.
It works be so much nicer if these blocks could render differently on mobile; it's surely not that hard for a computer to determine what is coffee and what isn't?
No. I'm fine zooming in or trying multiple times, I spend much more time trying to read which becomes harder when more and more of the screen is taken up by things that are not text.
I suppose it is easier to control the actual size, padding and so on with SVG. Unicode does have correct glyphs though and I suppose that the exact rendering of the arrows is not much of a priority. However currently they are blurry on any high dpi screen.
Contemporary design is overly focused on curb appeal. IMHO, it's an understated problem.
IMHO, the cause is a combination of commercialism and of visual designers being inculcated with inappropriate values and training from traditional print design.
> IMHO, the cause is a combination of commercialism and of visual designers being inculcated with inappropriate values and training from traditional print design.
I judge that it got a lot worse around the time the print-trained designers started to be pushed out of decision-making positions by the digital-trained ones.
I'm not sure I follow. Web design hasn't gone downhill over time - it's definitely improving, simply because the tools have allowed it do so. Perhaps you could provide some examples from the era of print designers and the era after them?
What I have observed has been that, while technology has allowed screens to become a more expressive design medium, visual designers tend to be content to cargo-cult techniques from the static, vastly simpler, and more expansive medium of print, and it shows.
I think we probably disagree about what good design is. I think that precisely
> static, vastly simpler
is the right way to go.
This is mere consistency on my part. I hated (most) Flash sites back in the day, and wasteful giant images, et c. I hate (most) JS/CSS cycle-burner sites and giant wasteful images today. Consistency is good. Shit not moving around on my screen is good. Controls working the usual way is good. Consistency, predictability, and clarity are good UX.
The basic 2-or-3-column layout with a thin header was fine, with maybe some tweaks for smaller screens these days. Early-period web's allergy to non-content large media files was good. Design mostly gets in the way, while making everything more expensive. I think its function was described perfectly by someone the last time we had a design thread: it's peacock feathers. Keeping up with design trends is harmful to function, but signals wealth (=stability, reliability, social proof of value) precisely because it's so costly and pointless. This doesn't mean it improves my experience on the Web.
Right, and this is the problem. Digital information is frequently neither. Denying this and shoehorning it into techniques that work for simpler, more static information is hiding the message for the medium. It's bad design.
>The basic 2-or-3-column layout with a thin header was fine, with maybe some tweaks for smaller screens these days. Early-period web's allergy to non-content large media files was good. Design mostly gets in the way, while making everything more expensive.
This wasn't because print designers were in charge. It's because the design industry, for the most part, is allergic to innovation and was ignoring the web. The web designers weren't designers - they very junior developers with none of the pretensions or preconceptions that come with formal training as well as all the problems that come with that, and a really primitive medium.
It's rather telling that returning to those pre-css days would improve things, is it not?
To each their own opinion, of course. But I think web design has significantly gone down hill. And I'm not even old enough to truly remember pre JS/CSS web.
Webpages these days are designed like casinos. I want specific content but they're keen on bombarding me with all the stuff they want me to want.
That's not a design problem though: that's a you're-not-really-the-customer-problem.
It is, however, exacerbated by designers who think they are creating fashion magazines. In addition to being co-opted by bad marketing, usability and function has, ironically, also been subordinated to immediate visual appeal.
However, you can't deny that we've some way since AOL, mirabilis and geocities.
One shouldn't spend 2 minutes blocking elements in a page that you would read hardly 5 minutes. That's counter productive and annoying and should be automated. An extension helps in that.
The picker creates rules that are persistent, so it would be 2 minutes over however many times you visit the site. Not too dissimilar to looking for the extension and installing it.
That's what I thought as well, but I repeatedly find myself blocking the same elements on Medium. Maybe there's some element of randomization in the ids of the offending elements, or else I'm using the extension wrong and only creating temporary rules somehow.
This extension, however, has the advantage that it checks for a <meta property="al:ios:app_name" content="Medium"> in the document instead of applying to a hardcoded set of domains.
This is awesome and will hopefully prod the developers at Medium to rethink the design (if your users need a Chrome extension to make your product usable, it's got some immediate problems that deserve a resolution). The sticky nav bars are really just terrible, and sometimes render parts of the page completely unreadable. It's surprising that these problems were not caught during basic usability testing. Thank you OP for fixing them!
I wish there was something I could do like this on mobile. I've given up trying to read these pages on my phone now, it's just too distracting to have all their UI clutter in the way.
What I don't understand is how they let it get this bad?
Firefox for Android has both reader mode which works fairly well with Medium, and you can install uBlock Origin, which declutters most mobile sites from ads (not sticky headers though).
Not sure how it got this bad, but there are ways to solve it for all sites (reader mode)
I use Safari's Reader mode to get rid of this stuff and modal dialog pop-ups asking me to subscribe or whatever they're asking. I also use Firefox Focus to remove ads on iOS or Adblock for Safari on desktop.
A bit off topic, but why does Medium add random numbers and chars at the end of their URLs like the example below[0]? That's annoying and distracting when referring from other platforms. Is that a git hash or something? If so, can I check revise history of the article?
I disagree, the action taken to create an extension and get it on HN sparks the discussion that can trigger real change. A boycott only works when you have a sufficient number of people doing the same, or in this case, a sufficient number of people bypassing the hostile UI.
Unfortunately lots of information is hostage to bad web design. I would really like to vote with my X button the web pages and apps of my uni, but instead I have to write scrapers to extract RSS feeds from their JS mess or else I can't know what's going on. Oh, the other option is to follow them on Facebook or Twitter. Also, the news sites often have hostile UIs, but one needs at least one of them.
In theory I agree, but reality makes that difficult to adhere to. To me, 15 minutes throwing a browser extension together is worth not being assaulted when I end up on a Medium-powered site. Same reasons I use an ad-blocker.
Funny thing is I avoid reading hackernoon blogs because of the green banner hurting my eyes. Not trying to be funny but I genuinely find it uncomfortable. So interesting that in the README file of this project hackernoon is shown as the example lol
Yeah, and as a contributor I have been surprised to see that they are now injecting (neon green) advertising banners into some of my articles. I don't mind it at the bottom, but I don't love it at the top.
You save any Medium article to Instapaper and read a readable version with your readability settings. That's what I'm doing now for stuff more than a few minutes long.
On a related note, I tried to follow the RSS feed for a few people on medium and every single comment by the author comes up as a new item, leading to a flood every time they're active on the sire.
If there was a way to get all of the comments to show inline per their thread on the main page that would be helpful.
Having to click a link, go to another page, have that page jump around on you and still need to click another link to show the replies is pretty terrible.
I'm waiting for "Show HN: How to blog with a static site generator". Medium is getting to be too much. If you're a programmer, at least use a static site generator (Metalsmith, Hugo, Hexo, etc.). You can deploy to Github pages with git or
via drag-and-drop to a service like Netlify.
Yep. That seems to be a trend with user content driven sites. They start off promising a clean, minimalistic look (like Medium, Imgur, Reddit, etc) and then gradually end up cluttering their site with more and more useless trinkets.
Indeed, it's so bad for image sharing sites that they seem to get replaced every year or two when the last one starts becoming more and more hostile towards their users' experience.
"Brands" generally follow the same pattern: grow the base, then cheapen the products to increase profits. Eventually the Brand will be replaced with the next big thing, or the Brand will survive long enough to provide a consistent product that people accept.
Website readability in general has degraded seriously in recent years. Flat buttons/links that can't be discerned as clickable, low contrast text, the resurgence of mystery meat navigation and plenty more annoyances I don't even remember.
Luckily we have better tools these days to combat them. The fact that firefox has a built in reader mode is indicative of how user hostile web readability is these days.
Man, those were the days. Though I guess it's managed to keep them at the top, perched on top of people who still think it's The Best Place to Host Your Content™
I for one am shocked, shocked, that independent publishers colluded with a single company in establishing a near-monopoly in hosted content delivery because of their "publisher-friendly policies" and said company is now taking advantage of that position.
(in my humble opinion) they should just as much look at the content they publish as how they publish it. The amount of clickbait quality articles on medium is sky rocket high.
This has slowly been driving me insane. I will for sure use this. I'd think the bars would be common criticisms and medium would have done something to limit them by now.
Does anyone have a problem with Medium on Firefox? For me, it doesn't have the top and bottom bars. I've switched uBlock off and this is still the case. Is all of this extra chrome, a Chrome only thing or some A/B testing?
I use Just Read[1] for things like this. It attempts to pull out the relevant text and put it in a user editable style sheet (the default is fine for many).
You'll get a different (less annoying) experience if you're logged in. However if you haven't logged in on a Medium site (these [0][1] might be some), you'll get the annoying sticky bars.
Is there anything like this for Gmail? The two large sticky bars (the bar where the search box is on, and the toolbar bar) are wasting too much of my screen estate. Thanks in advance.
It's really sad to see Medium turning into this. I used to like Medium, now I avoid any site that uses them (which is getting harder as they now allow custom domains).
OP mentions a similar platform he made, Write.as, in the README. It doesn't do as much as Medium (I think that's on purpose) but they have a small Medium-like network for authors there [0].
I've been on it for over a year [1] and really enjoying the simplicity myself.
My short list would be a self-hosted blog (static site), Reddit (what I'm actually using), Wordpress.
Github has some blogging options that I've been meaning to look at, and might also be an option.
I've been looking for a better solution, and Medium have pretty much elected themselves out of consideration with their recent directions, styling and otherwise. No confidence.
Static website generators and GitHub pages, WordPress.com, Blogger, typical hosting sites. I think some forums and Reddit style link aggregators have places for longform content too.
I was considering my own direct alternative to the site, but at the moment I have other things to work on.
AFAIK Medium's selling point is the network, i.e. the easy access to potential followers. My website is static via org-publish, but nobody knows it and if I wanted to popularise it I'd have to do many things. With these networks+hosting services like Medium and Tumblr all you have to do is put some tags. I was asking for this sort of alternatives.
Then I'll say that I'm working on something similar, it's just not ready at the moment. I'm retooling my gaming site into a gaming alternative to Medium and the likes, and will hopefully launch a general version for all subjects at a later date.
It seems to be commonly regarded that Medium has the most excremental reading experience of any publishing platform.
Remember when they used to publish blog articles about making the web a better reading experience? That seems a bad joke now.
So Medium have trashed their brand. But my question is: do they know? Do they care? Do they read all the articles about how appalling the reading experience on their platform is? What's going on?
I don't work for Medium, so total speculative fiction here, but let me imagine the life of a product manager at a small growing company...
Somewhere at Medium, perhaps even reading this thread, there is a Product Manager in charge of, among other things, making the Medium reading experience great. Unfortunately, he (not knowing his gender, just picking a pronoun) is angry and frustrated every day. He wants to do right by users, indeed, his mission is to make everything about the product awesome. But, there are so many competing priorities and agendas. His boss will say things like "We can't measure whether it's a better reading experience, so your KPIs are things like conversion, reach, number of user engagements!" So he grudgingly sets out to optimize these things, while at least trying to keep the site usable. His first suggestion is to pin all the site links to the top of the page so people are always clicking around Medium, churning from page to page. The "Social Sharing" PM reaches out with "Hey, buddy, why not make the social sharing buttons twice as big, and never remove them from the page--both our numbers will be through the roof! The "Audience Growth" PM wants to make sure there are at least 3 signup buttons visible at all times, because obviously users are just not seeing them. He wants to help his colleagues, and, he is expected to drive his KPIs. We can compromise a little here, and a little there. All these internal stakeholders will be grateful, and he will be seen as a team player.
So he has the developers go off and implement these things, rolls them out, and checks the all-important metrics dashboard. Lo and behold, all his targets are met! Success! A few more stock options next review-time! But in the back of his mind, the product owner in him is screaming in horror at what he's putting his users through in order to meet these goals. Maybe he even brings up these concerns with his management, but they dismiss them and want to instead talk about the next floating button bar he's going to add to drive those numbers even higher...
Medium's trashed their brand way before this. Likely the point where quality content became less of a concern than 'anything with words' and the staff decided that promoted the vapid and overtly political was more important than finding decent stuff for their users to read.
Of course, all these layout changes and annoyances have helped even more, so what was once a minimalistic site with a decent quality control standard is now just any old blogging platform with any old blogging platform's ugly awkward to use layout.
Readability seems to be in regression, and not just at Medium. If a (near-) optimal solution has been found, and you are committed to doing something different, it is likely to be suboptimal.
> Do they read all the articles about how appalling the reading experience on their platform is?
Doubtful
>What's going on?
They don't know how to monetize any better than any other web site. They're following what other sites do to try and engage the people who do visit Medium longer. The problem is, what works for sites where the reading experience is secondary doesn't work for a site whose reading experience is primary. Somehow they either don't know that or cannot think of anything better to try.
They're probably seeing an uptick in signups, subscriptions, and all the other metrics they care about---so they keep going down the rabbit hole of adding more modals and call to actions, and popunders. They don't realize at the end of that track is a brick wall of 0% growth as new visitors are instantly turned off.
If I werent' already committed to using Medium, I'd click away the instant I saw the site as it reminds me too much of spam-blog pages.
I just reached the point of "ew, no" when I'm about to click a link that I can see is going to go to Medium due to the popups that start, "You've read N articles on Medium, ..."
I can't recall what the rest says because by the time I've scanned the first part I've already right-clicked to "inspect" to delete the element. (Then I cleared all Medium cookies, but that doesn't prevent them from showing the popup again, and it still has the N so obviously they are storing it server-side. :-( )
The formula works! If it can get that guy in the White House, it can sell anything.
Also, I think it's kind of delicious when a variant like this is used to actually describe a previously "Y/Great" thing (Medium was readable at some point). Kinda hijacks the meaning from the original phrase and makes it less shitty.
But really, any sort of thing that blocks, as the Github README says, "annoying, user-hostile software", is a good piece of software in my book.
[0] https://alisdair.mcdiarmid.org/kill-sticky-headers/