5 ms·
It’s not the designers’ fault, it’s developers seeing everything as a technology problem rather than a user needs problem (and I’m speaking as a developer). The
by robin_reala 7y ago
It’s not the designers’ fault, it’s developers seeing everything as a technology problem rather than a user needs problem (and I’m speaking as a developer). The best solution for any given problem is usually the simplest, at least until you find that you specifically need extra complexity to solve a user need. Unfortunately modern client-side development is fixed in a mindset of developer efficiency above all, which doesn’t naturally lead to simple solutions.
- imustbeevil 7y agoEven as a tech problem, the new UI puts less information on the screen and makes it harder to interact with. It's worse by every observable metric, and I can't imagine they have successful A/B testing, because every other site on the internet that does a redesign (Facebook, Twitter, Youtube) commits to it, and Reddit has allowed users to stay on old reddit for like a year now. The weirdest thing is that successful alternative reddit UIs exist, so all they would have to do is incorporate the features people are using alternative UIs for. They did virtually the opposite of that. This is exactly what Digg did, and the solution for Digg wasn't "allow people to use old Digg", it was "Everyone left and went to reddit".
- slig 7y agoThis time, though, Reddit doesn't have a suitable competitor and they know it. Power users will whine, but they don't have anywhere else to run.
- zzzcpan 7y ago> A/B testing Despite popular belief, A/B testing doesn't actually work for usability testing. So successful A/B testing usually means bad usability practices.
- acoard 7y agoCan you provide some evidence or reading on this? I understand A/B typically measures _conversions_, but couldn't you have quantifiable proxies for usability? For example, things like form abandonment and time-to-complete a task. To be clear, I'm not contesting the value of usability tests involving watching actual users. Our UX team does this with every release and it's invaluable. But I would think data from A/B testing would only help inform our decisions.
- cco 7y agoThe A/B testing would only be concerned with ad click through and time on site, by those metrics the redesign could be a success, I have no idea.
- amelius 7y ago> It’s not the designers’ fault, it’s developers seeing everything as a technology problem rather than a user needs problem Yes, but in a big company developers don't define the product, (product) designers do.
- chronogram 7y ago“Big company” developers: do you call yourself “Unit 3829” or “autonomous drone 283”? Because seeing comments like above so frequently, you’d think developers are sacred and never do wrong but have no free will or input.
- QuickToBan 7y ago> you’d think developers are sacred and never do wrong but have no free will or input. Developers who express their free will strongly risk getting fired. It pays to keep their mouth shut and let the firm torpedo itself.
- amelius 7y agoWhy else do you think the industry favors young developers over seasoned ones?
- Spooky23 7y agoFunny you should say that, I just had a fun email exchange with the leader of Procurement Team 13-T.
- sizzle 7y agoCorrect, but developers are directly responsible for the performance and stability of the platform. I highly doubt the designers specified in their designs that the website should constantly display: "Sorry, we couldn't load posts for this page" as mentioned in the OP. If we are talking about poor interaction design patterns and unintuitive conceptual models with no prior usability testing/UX research then designers/researchers deserve the blame (assuming devs didn't botch the implementation and create usability problems), but this isn't the issue at hand... platform stability is and that's the job of the developers who code it.
- vcavallo 7y agodoes “simplest” here mean simplest for the user to use, or simplest for the dev to build? these are sometimes completely opposed to one another. an example: filtering a large table of data on multiple fields in an otherwise fully-server-rendered project. doing it server-side is likely mostly sql queries using existing stack and re-rendering the page after some “submit” click. doing it client side means the data can react immediately to selections and never require a full page refresh (which might be costly, depending on the rest of the page). the dev downside is that this might require an enormous amount of new stack and complexity.
- zzzcpan 7y agoI don't think it has anything to do with developer efficiency. My impression is that client-side development is seen by developers as sort of an unimportant throwaway specialization. Nobody cares about it or bothers to learn any of it even to be able to produce something of quality. First thing people stumble upon is the thing they decide to learn, which almost always means popular javascript framework of the month or whatever. Progressive enhancement, UX, accessibility are all unreachable expert level things to them. I guess my point is there are simply almost no skilled client-side developers available for hire for large organizations to be able to do well on this.
- robin_reala 7y agoAccessibility and user experience are base level skills for calling yourself a front-end developer. If you don’t have them then the products you build will be unusable and illegal. Or, the put it another way, full stack is a reformulation of jack-of-all-trades.
- basch 7y agoresponsive sites and mobile apps have led to the mobilification of desktop sites. there is absolutely a branding/marketing/design trend pushing shiny over usability. ive noticed it in banks in the last 5 years, where they have replaced dense powerful beautiful text based interfaces with giant versions of their mobile apps. huge buttons, minimal screen density. https://m.signalvnoise.com/why-i-love-ugly-messy-interfaces-and-you-probably-do-too/ https://m.signalvnoise.com/why-i-love-ugly-messy-interfaces-... i liked the old reddit BECAUSE it was a simple list of links that all behaved exactly as a link should when i click it, and didnt attempt to put visual flair over usability.
- wott 7y ago> in banks in the last 5 years, where they have replaced dense powerful beautiful text based interfaces with giant versions of their mobile apps. huge buttons, minimal screen density. You forgot to mention that they load progressively and you never know when the page layout is done and stable, so when you think you click on something, your click ends up being received by a button/link which popped up in between, and there we go, launching a new page or performing an unwanted action, and then going back to the previous page (well, you've got to find the way to go back, because of course it may or may not obey the standard "previous page" command, or it may or may not be a one-page app with a specific way of going back [a cross, a link, a button, ...] placed in a specific place [upper right corner, upper left, bottom row, bottom left corner, bottom right, in a window, in the full page space, etc.]), and waiting for the original page to reload completely one more time...
- basch 7y agoand every bank seems to think its own quirks arent a big deal, but when you have 6 different apps, each with its own quirks; remembering how the back buttons work is a nightmare. citi is especially bad, but so is just about every bank when you switch from your account to cc rewards.
- Xelbair 7y agoI would rather say that it is a management problem. Focus shifted from building userbase and sustaining business to milking ad revenue.
- ishjoh 7y agoSetting aside modern FE development practices, are you trying to say that the redesign of the reddit UI and it's poor usability is due developers choice of tools instead of what designers created?