denise: Image: Me, facing away from camera, on top of the Castel Sant'Angelo in Rome (Default)
Denise ([staff profile] denise) wrote in [site community profile] dw_suggestions2010-01-05 09:26 am
Entry tags:

Why more options are generally bad UI

One of the most frequent things I see in [site community profile] dw_suggestions is a pattern that goes like this:

User A, making the suggestion: I don't like this behavior foo, and I think it should do bar instead.

User B: I rely on this behavior foo, and if it did bar, it would break my use of the site.

User A: Okay, so, I'd like to refine my suggestion to instead propose an option so people who want foo can have foo, and people who want bar can have bar.

(Or, the compact version: user A, knowing that their friends use foo but they want bar, proposes a suggestion that jumps straight to the option.)

Those suggestions are less likely to get accepted for implementation. Very, very, very much less likely to get accepted for implementation. Why? From a user interface design standpoint, generally speaking, options are bad. They force someone to make a choice, and nine times out of ten, those choices aren't necessary.

Generally speaking, people like to feel like they've customized their environment to their own tastes -- but, conversely speaking, people also feel uncomfortable and awkward when they're offered a choice they don't understand or can't "intuitively" figure out.

Now, for the purposes of web usability -- hell, for the purpose of computer usability, period -- "intuitively figuring something out" is defined as "it does what I think it should do, based on my collected history of experience with computers". This generally means, "it follows the usage model that I'm familiar with" -- which means, in most cases, "it works the way the market leader works", since that's what people are likely to be most familiar with.

A new user, confronted with a panel of options, won't read the explanatory text or the FAQ or the carefully-crafted description of each option. They'll look at the options and decide that this software is too complex and too hard to use, and they'll go back to using what they're familiar with.

Too many preferences make people choose no preferences at all.

As an example, let's take a look at the Preferences panel of Scrivener. Now, don't get me wrong: Scrivener is, hands down, the best word processor available for anyone who has to do writing, research-wrangling, or anything involving creative work. I love this program. I adore this program. I am writing this entry in this program right now.

If I accidentally lose my preferences on this program, or if I'm setting it up on a new machine and don't have my preferences file at hand, it takes me forty-five minutes to rebuild them. Why?

Here's the preferences tab you see when you open the program's preferences:

screenshot of preferences pane with a hodgepodge of about 20 different options, most of which are confusing or otherwise not immediately essential, marked 'General'

Now, mind you, that's just one pane. There's also the text editing pane. And the typography pane. And the fonts & color pane. And the navigation pane. And the full screen pane. (Which does not put the program into full screen mode, of course -- it merely governs what happens when the program is *in* full screen mode.)

Don't get me wrong. I love Scrivener. I love it to death. It's the best program for managing creative output on the market, and at least half of that is because of the various options. Writers are cranky, cantankerous creatures -- I know because I am one -- who will endlessly fuss with their work environment until this window is precisely pixel-centered, and that window is just the right shade of blue, and claim that they're unable to work if they can't get it to look exactly the way they need it to work. (I have been known to engage in more than a little bit of this redecorationism myself. Let us not ask about what I needed to do before I could get started on this essay.)

Scrivener's smart about their eight gajillion options: it allows you to save and load preferences files, so you can transfer them from computer to computer, recognizing that it's a lot to ask people to set each and every one of their preferences. People still take one look at the preferences pane, throw up their hands, close the pane, and accept the defaults. And if you don't lose them as users then and there, it will take them months, if not years, to figure out all the things your program can do. (True story: While taking those screenshots, I figured out at least three options that I've been wishing Scrivener had. It did. I just couldn't find them. And I am, let us say, a less-than-typical user. If I can't find them, my father has absolutely zero chance of finding them.)

Open Source software has a horrible reptuation for usability.

You guys have seen Warmouse's OOMouse that was just announced, right? With the 18 different mouse buttons, most of which are smaller than the average adult human being's fingers? Contrast that to Apple's mouse, which has no mouse buttons: it uses touch-sensitive technology to figure out where you're clicking and what you want to do with it. Which one of those products intuitively looks more usable?

The response from most of the tech world has been laughter, but it's a real symptom of a major problem with open-source software: the "all things to all people" approach, where instead of making decisions to prioritize one option or the other, the project maintainer chooses both:

Worse, given that peer esteem is a crucial incentive for participation, deletion of functionality in the interest of benefiting the end user creates a strong disincentive to future participation, perhaps considered worse than having one's code replaced by code that one's peers have deemed superior. The project maintainer, in order to keep volunteer participants happy, is likely to keep functionality even if it is confusing, and on receipt of two similar additional functionalities, keep both, creating options for the user of the software to configure the application to use the one that best fits their needs.
We speculate that freedom of choice may be considered a desirable attribute (even a design aesthetic) by many OSS developers. The end result is an application that has many configuration options, allowing very sophisticated tailoring by expert users, but which can be bewildering to a novice.

(That's from a fairly scathing condemnation of Open Source usability: The Usability of Open Source Software by David Nichols and Michael Twidale.)

The people who are most likely to be open-source developers -- and, to some extent, the people who make suggestions about something an open source project should implement -- are the expert users. Their idea of how an application should behave is entirely different than J. Average User.

The traditional hallmark of J. Average Userdom is often heard as the "mom test", the "grandma test", or the "girlfriend test" -- if you gave this software to your $person, how well could they use it -- which I dislike the phrasing of because of the inherent sexism while at the same time liking the motivation behind it. Personally, as I mentioned a few minutes ago, I use my father as the watermark. He can't find anything if it isn't saved to the desktop and needs written instructions (that he keeps next to the computer) to perform any task more than twice. He's a brilliant man, mind you. His mind just doesn't work in the same metaphors that software developers consider a matter of course. So, when he sits down to the computer, he looks at the 18 mouse buttons, or at the hundred-and-fifty-something different preferences, and he gets annoyed: why are you asking me about this stuff? I'm just trying to write a letter.

You have to balance features/options with usability.

Now, I'm not saying that all options are bad. For one, options are generally hallmarks of features, and features are generally good things. There's a generally-accepted truism in the software world that 80% of your users will only use 20% of your features, but there's a corresponding truism that each person will use a different 20% of your features. You can't just cut out the other 80% of features and skate by, because everyone's usage pattern is different.

Correspondingly, there will always need to be some options. Let's say you've identified two major use cases for your software, equally divided: 50% of your users are Case A, and 50% of your users are Case B. You're about to release a feature that will make life 100% easier for Case A, but seriously impede Case B. It'd only be smart to allow Case B users to choose not to use that feature. In most cases, the answer will be "well, if it makes your use of the software harder, don't use that feature", but in some cases, in order to implement the feature, you need to change some of the core ways your software behaves. In that situation, it's logical to offer an option to turn the feature off: if you don't, you might be picking up probably about 20% more users in Case A (because of the 80%/20% rule) who are attracted by the feature, but you'll be losing the 50% of users in Case B who suddenly can't use your software to accomplish what they've been using it for.

(This is a dramatic example. Real life examples of software design are a lot harder and a lot more subtle.)

But you also have to take into account the users who will take one look at your product and wander away because they think it's too hard to use. These people are nearly impossible to measure, because they don't show on any stats or metrics. (On Dreamwidth, we can measure them a little, by how many people have accounts that haven't been active in the last 30 days, but that's not the best metric, because there's no way of knowing the reason they haven't been active, and there's no way of counting the people who never created an account in the first place.)

So, options are a tradeoff. Every time you add one, you're asking the user to make a choice, and you the software designer are promising them, implicitly -- by the very fact that you are offering them a choice -- that this particular choice is important. It's major. It's something they should care about, because you care about it, enough to tell them that they should have a reason to pick one or the other. If you didn't care about it, you'd just pick a default and set it for them, right?

This means that if you're going to give the user a choice, it needs to be important enough to justify that choice.

So, what do we-here-at-Dreamwidth consider important enough to justify adding options?

To cycle back to the beginning of this, the discussion about [site community profile] dw_suggestions, what sort of "add an option" suggestions are likely to get through to the point of being implemented?

At this point, and until we can get our options tabs redesigned and cleaned up -- which is an ongoing project of mine -- probably few to none. The biggest ones we're likely to add are things that have a very good argument-in-favor that rely on one of two things: things that improve privacy or things that improve accessibility.

One thing we haven't done a lot of, but that I'm way less opposed to (and Mark is even more anti-options than I am, but he's also more okay with this) are things that remember the last-used-value. For instance, one of the common things we hear suggestions for: "let me pick what should be the default in the search bar drop-down". (Should it be Site & Account? Should it be Interest? Should it be 'this journal', when we add it? etc.) That's such a minor thing that we'd likely never add an option for it. But remembering the last used value, so that if you always use it to search for Site & Account and never Interest, you only have to change it once? That's a small touch that improves usability for people whose usage pattern doesn't match the majority, while not disturbing those whose usage pattern is the majority and not adding another option to look scary to people. (And it's something we'll be adding as soon as I figure out how to do it!)

So, if you're looking to suggest a new option to do something, stop and ask yourself two things:

1). Is this option something that a majority of users care about?
2). Is this an itch that can be scratched with a different solution that doesn't involve adding an option?

If the answers are, respectively, 'yes' and 'no', then go ahead and suggest it. If not, start brainstorming about what those different solutions might be.

If all else fails, post a suggestion identifying the problem and saying that you don't think this is enough to warrant a new option, but you'd like to start some brainstorming on what other solutions we might come up with. We've got a lot of very smart people who can come up with different solutions once a need has been established, and Mark and I are both perfectly fine with using [site community profile] dw_suggestions to conduct that brainstorming.
cesy: "Cesy" - An old-fashioned quill and ink (Default)

[personal profile] cesy 2010-01-05 03:12 pm (UTC)(link)
What I still don't get about this, is why you can't have an "Advanced Options" section, that will be ignored by the people who get confused by too many options, but so it's still there for the people who want it. I mean, I get the basic argument, but surely that only applies to options in the main options section? So long as it's clearly labelled as Advanced Options or something like that, I don't see why you can't stick a load of extra things in there.
pseudomonas: (Default)

[personal profile] pseudomonas 2010-01-05 03:27 pm (UTC)(link)
I think this depends on the community structure, especially as regards support. If someone wants to do something that other users are doing, either telling them "it's possible, but you're not clever enough" or having them try to sort it and fail, may be worse (in their case) than something not being possible.

I guess in DW there are complicated things like custom styles that people can fiddle with and mess up on their own, though, and it'd be hard to imagine a set of options more complicated than that lot.
zvi: self-portrait: short, fat, black dyke in bunny slippers (Default)

[personal profile] zvi 2010-01-05 04:00 pm (UTC)(link)
In part, because Advanced Options are a Random Collection O' Stuff. Random Collection O' Stuff is terrible: it's harder for a user to find that an advanced option exists, it's harder for a developer to update things that live in multiple places (and presumably there would be advanced options that were related to regular options, and vice versa), and it provides a perfect platform for feature creep (i.e. adding features that no one actually wants just so one can say one has more features.)

Also, what makes an option "advanced"? This is posting your thoughts to the internet, it's not rocket science. Very few of our options involve concepts that the average user would really have difficulty understanding. (Styles is about the only area where I think things are largely unintuitive.) It's the sheer number of options that are overwhelming.

Also, advanced users also get overwhelmed by choice. They won't stop using the product, but they also will not set preferences, if presented with too many choices.
aveleh: Close up picture of a vibrantly coloured lime (Default)

[personal profile] aveleh 2010-01-05 04:47 pm (UTC)(link)
I like software that does it by having "advanced [subset] options". That is, you have Options, with tabs A-F, each of which shows maybe 5 options, and at the bottom of each of those options, you can expand the remaining options for that tab.

(As opposed to having tabs A-F plus tab G for all the advanced options, I mean.)

Then, you can still find that option in the right place if you're looking for it, but you also aren't overwhelmed by too many choices.
zvi: self-portrait: short, fat, black dyke in bunny slippers (Default)

[personal profile] zvi 2010-01-05 05:02 pm (UTC)(link)
Yes, that's a better solution to the random collection 'o stuff problem, but it doesn't address all of the problems.

More options means having to maintain and support the software in more potentially different states. More options means more code. More options means refactoring the menus whenever you had an option :(

Or, in the simplest analysis, more options makes it harder to code, design, and support the software, while not actually offering any benefit to most users.
aveleh: Close up picture of a vibrantly coloured lime (Default)

[personal profile] aveleh 2010-01-05 05:09 pm (UTC)(link)
That's why I didn't argue with that part :) I'm just saying that in cost vs benefit, *where* to put it doesn't have to be as big of a cost.
elf: Rainbow sparkly fairy (Default)

[personal profile] elf 2010-01-05 04:13 pm (UTC)(link)
It ties into the same problem--people will request a feature, and be told "we have that! It's in the Advanced Options area, along with 1,342 other options you *don't* care about!"

You wind up pleasing the 3% of the userbase that adores tinkering with fine settings ("OMG, is this where I set the kerning on my custom javascript-inflicted fonts to 84.3% of normal? Yay!"), confusing 50% of the userbase (who might love a button that inverts the colors of a journal so they can switch from light-on-dark to dark-on-light but don't want to have to change the whole layout to ?style=mine), and annoying the other 47% who were trying to figure out how to search by location, wandered into the wrong part of the settings, and now their journal entries put all their italic text in 14-pt Impact in red, and they can't figure out how to undo it without losing all their other settings.

Putting the choices in a more restricted area doesn't change the essential issue, which is that people are more frustrated by choices they don't know how to use efficiently than by not having choices.

Hiding those choices (in a special "advanced options" area) avoids some of the frustration by making more people think those options just don't exist, but it adds annoyance in a new area, when they discover the option they've wanted for months or years exists, and nobody told them.
ninetydegrees: Drawing: a girl's face, with a yellow and green stripe over one eye (Default)

[personal profile] ninetydegrees 2010-01-05 04:13 pm (UTC)(link)
I think I'm with you on this. I'd rather have one or two options I consider vital and the possibility to gleefully ignore all the other ones I have no use for/don't understand that a program with less but better organized options. In the end, options are often the criterion which makes me choose between dozens of similar programs.
azurelunatic: Log book entry from Adm. Hopper's command: "Relay #70 Panel F (moth) in relay. First actual case of bug being found" (bug)

[personal profile] azurelunatic 2010-01-05 04:29 pm (UTC)(link)
My thoughts, which as always are fairly random, and this time aren't even based on extensive study, is that options to make something 99% perfect for a subset of the group after twiddling are likely to lead to the default being not exactly unusable, but unused, neglected, and that dreary good-enough or barely-good-enough that one is familiar with from countless OSS apps where the programmer was also the designer, having never studied good principles of design.

Then, once some time passes, the default may start to get that unmaintained look, where what was once cutting edge design and features are getting outdated and show it. Meanwhile, since the power users have been using customized settings all the while, and are perfectly happy with the user experience they've crafted for themselves, and have been suggesting new features based on what they've been working with (not with what the defaults are), there's still updating, but it's going on in the weird edges, and someone who comes in is still going to be stuck in 2008 or 2001 or whenever.

If you're the power users and you're using the default, or close enough to it, and your devs and designers are using the default, you're going to be noticing the obnoxious things about the default and getting them fixed.

... you know, it might be worth throwing some research at that, to see what falls out. (Before I do anything like that, though, I need to finish that other thing.)
pseudomonas: My rat is confused by technology (technology)

[personal profile] pseudomonas 2010-01-05 04:37 pm (UTC)(link)
The default thing gets to be a bit of an issue in sites where things get optimised for logged-in users, even though 90% of the views are from logged-out users - since it's the logged-in minority doing the customizing.

(Also, having too many options makes support hell. "You just click the button that's in the top bar, unless you chose the option to put it in the sidebar, or to make it a drop-down menu. How do you find out what options you chose? You just click the button...")
Edited 2010-01-05 16:37 (UTC)
sophie: A cartoon-like representation of a girl standing on a hill, with brown hair, blue eyes, a flowery top, and blue skirt. ☀ (Default)

[personal profile] sophie 2010-01-07 06:49 pm (UTC)(link)
(Also, having too many options makes support hell. "You just click the button that's in the top bar, unless you chose the option to put it in the sidebar, or to make it a drop-down menu. How do you find out what options you chose? You just click the button...")


It's also a big reason why DW uses the same menu in its different site schemes, and in fact the menu only has to be changed once by a developer for it to be changed in all the site schemes. That in turn (well, the idea) lead to multiple card sorts being done before DW even went into closed beta to figure out what exactly was the best menu structure, since the LJ structure really didn't make a lot of sense.

Granted, there are some differences between web apps and local apps in regard to making choices; while it's true that virtually nobody is going to spend the time to customise their menu and toolbars in Word beyond the second or third time (if that), web apps are slightly different in that your choices can stay with you, no matter what computer you're using. But in the case of DW, a lot of users will also be creating new accounts, and you have the same problem of needing to configure everything all over.

So, yeah.
trinity_clare: (Default)

[personal profile] trinity_clare 2010-01-06 02:15 am (UTC)(link)
...Excellent point.
kyrielle: A photo of kyrielle, in profile, turned slightly toward the viewer (Default)

[personal profile] kyrielle 2010-01-06 05:00 am (UTC)(link)
Hmmm. So another approach that might work is identifying suites of settings (if you want A set to 1 you usually want B set to 13 and so on) and tying them together behind one option, IF we can find things that tie together.... Maybe, anyway.
Edited (Seriously, how do I manage to accidentally hit tab-space, thus pressing the submit button, when trying to type an a? How do I do these things?) 2010-01-06 05:01 (UTC)
ninetydegrees: Drawing: a girl's face, with a yellow and green stripe over one eye (Default)

[personal profile] ninetydegrees 2010-01-06 02:29 pm (UTC)(link)
Re: your edit
This happens to me all the f* time. ;)
triadruid: Apollo and the Raven, c. 480 BC , Pistoxenus Painter  (Default)

[personal profile] triadruid 2010-01-06 04:12 pm (UTC)(link)
I disagree with this approach and mindset entirely, but I'm sure that's because I'm part of the "problem user set" that actually uses options, customizes my toolbars, and otherwise makes myself unsuitable for mass production.

Oh well.
kyrielle: A photo of kyrielle, in profile, turned slightly toward the viewer (Default)

[personal profile] kyrielle 2010-01-07 04:52 am (UTC)(link)
Microsoft ribbon bars. *weary* Oh, if only they contained what I wanted in some place that made sense to me.

I miss my customizable toolbars so hard.
triadruid: Pseudocode for "If nothing else, remember this." (codemonkey)

[personal profile] triadruid 2010-01-08 05:01 pm (UTC)(link)
I'm putting off "upgrading" to Office 2007 as long as possible - in my org, that means the end of February 2010. :(
liv: Stylised sheep with blue, purple, pink horizontal stripes, and teacup brand, dreams of Dreamwidth (sheeeep)

[personal profile] liv 2010-01-08 11:29 pm (UTC)(link)
Me too, definitely. Whenever I try out a new service, the first thing I do is customize lots of options. If I can't do that, I am far more put off than I would be by finding the UI confusing. And if I continue to use the service regularly, I get a lot of pleasure from tweaking. I suspect DW is atypical; people who want everything to just work are probably using Facebook or Blogger. A lot of people like LJ based sites precisely because there are so many options and they can make their journal look exactly how they want it. That's got to skew the user population heavily.

That said, if it's a case of introducing a feature, I think it makes most sense for the "option" to be simply not to use the feature if you don't like it. Features that are not core to using the site shouldn't be forced on users, either because the pathway is set up so you can't avoid them, or because there are CProds and other intrusive, repeated reminders about new features you haven't tried yet. This makes things much better for both the users who like everything just so, and those who prefer a streamlined UI.

I have no interest at all in using a GUI editor for HTML, I'd much much much rather hand-code (see under: atypical user!) But I have no problem with that option existing on the site, I just choose not to use it. But I do have a problem with, say, an update page that relies heavily on JavaScript and doesn't work in old or text-only browsers. The current system, where it just remembers your last choice, is probably preferable to having a ticky box with an incomprehensible title in a tab I can't find, but IMO the ticky box would be infinitely preferable to forcing everybody to use the RTE.

The other reason I don't completely buy this philosophy is that some minority users are not merely more fussy than average, but working round limitations, either technological or personal / medical, in the way they use DW. In many cases, options are an accessibility issue, not just a trying to please a wide range of fussy users issue. Since I really admire DW's deep commitment to accessibility, I hope it will be applied to this question as well.
lorax: Hardison is a Geek at Work (Lev - Hardison "Geek At Work")

[personal profile] lorax 2011-02-26 09:15 pm (UTC)(link)
I found the whole of this post very interesting, and while I don't personally agree with it, I think it probably is very valid for most users, I'm not sure it applies as much to DW users.

Most DW users are not coming to an LJ-style site for the first time, or entirely unfamiliar with the options and standards from other sites. The greater range of options, accessibility, and the philosophy of the site are what usually draw people over.

Like some posters above, I hand-code without using the editor, but think having an editor as an option is a good thing. I LOVE futzing around with options, even if I don't necessarily understand them at first.

Too complex to code, or adding in too much complexity into the system, thus making it impossible to support, is something that I understand, but the idea that too much complexity/too many options will turn away DW users doesn't really seem quite as valid. DW is also known for its accessibility commitment, which by definition seems to require a lot of options, because of the wide range of issues with use that are addressed.

And just as a random thought - Facebook (which no one is interested in DW becoming, so that's not what I mean, I swear!) has a million privacy options and is an arcane BEAST to navigate to turn them all on and off and figure out, and has been widely and horribly criticized for it, but judging by the sheer massive omnipotent taking-over-the-internet power of it, I don't know that it turns off people, except for people who PREFER to have more options for protecting privacy, and are thus wary of Facebook for not being protective enough - and having lots of options and ability to customize and privatize things on DW would be something those kind of users would want, I'd think.

This is all just comments from an uniformed user though!
matgb: Artwork of 19th century upper class anarchist, text: MatGB (Default)

[personal profile] matgb 2011-02-26 10:03 pm (UTC)(link)
Most [current] DW users are not coming to an LJ-style site for the first time

Corrected for you there. There's an important aspect to address within all of our suggestiongs filtering discussion, including this one, and that's the very simple point that, medium term, DW needs to outgrow LJ as being the source of new users.

I already know people that have DW accounts but have never touched LJ. One of them is a good friend and will be moving in with us soon.

What LJ users expect isn't a good thing to draw from, for two reasons, first being that we need to recruit people that aren't from LJ.

Second being that, for a number of reasons, LJ's usability sucks mightily. Denise was involved before she left in The Great Usability Project that some of us followed fairly closely when we could. The dropoff rate for newer users to the site was scary, people would sign up, go to do something,t hen give up, etc.

Making a website easy to use and non scary is important.

I like the way DW keeps most of the frontend fairly simple (although I think there are too many options in some areas and not enough in others), but I also like that, under the bonnet, I can hack away to my hearts content (and get help from site admins when I do something really stupid and/or obtuse).

But 'options' for a lot of features are bad in some ways and can lead to weird affects: I've put a lot of work into my journal's CSS so I can do display based tricks in post, but if someone views an entry of mine using those tricks in a different layout, some of them, well, don't degrade gravefully.

A typical DW user is, I suspect, always going to come fromt he 'power user' end of the scale, but that's a typical user. I want my friends to be able to use the site without getting horribly confused. Some of those people have problems turning their mobile phones on (seriously, double first oxbridge degree but fins the on switch confusing).

Sometimes, options are essential, othertimes? Not so much.
lorax: D is For Lots of Things (Sandman - Endless "D Is For")

[personal profile] lorax 2011-02-26 10:21 pm (UTC)(link)
"Most [current] DW users are not coming to an LJ-style site for the first time"

Corrected for you there.

I didn't meant to generalize, and apologize if it seemed dismissive. Growth is essential, from former LJ users and from those who've never been near it. I more meant that the majority of DW users from what I've observed (and thus a limited viewpoint, naturally) are usually people who have already used blogging sites of some kind. My wording could have been better.

Sometimes, options are essential, othertimes? Not so much.
I included that more to say that I was answering as a user, not as anyone experienced with anything but being a user of the site. The post was linked from the news post, which was why I wandered over and read and offered opinions in the belief that they were looking for feedback on the various posts mentioned in the news post. If that wasn't the case with this post, then I apologize. I have no illusions that my views are, or should be, the deciding factor in anything.

However, that comment sounded a good deal like "you're wrong, so be quiet", and was a little harsher than warranted, I think.
lorax: Big Dang Hero (MdMn - Middleman "Big Dang Hero")

[personal profile] lorax 2011-02-26 10:34 pm (UTC)(link)
This post is more informative than seeking-opinions, really!
My apologies then. :) Feel free to delete if you'd like to keep the thread cleaner.

Also, you mention accessibility -- remember, too many options is also an accessibility problem! Non-neurotypical people and people with anxiety issues (especially technical anxiety) are often stressed out to the point of panic by too many choices.
That's true, and a point I hadn't thought of!
matgb: Artwork of 19th century upper class anarchist, text: MatGB (Default)

[personal profile] matgb 2011-02-26 10:32 pm (UTC)(link)
That wasn't me being harsh. If that's how it read, apologies, but the opposite is true: feedback is always good. I subscribe to comments on Suggestions because it gives me a very broad view of the viewpoints of the site users (I'm very aware I'm an unusual edge case user so need to think through how others want the site to act).

the majority of DW users ... are usually people who have already used blogging sites

Yes, true. But that's because most of us transferred here directly from LJ for a vast variety of reasons (one of the recent exoduses was prompted by a badly implemented feature I'd been lobbying for, for example).

But that's current users, and we always need to be careful to not design for current users; it's potential users we need to attract, and that's a vastly different use case. And much more diverse.
lorax: Dark Side Roxors! (Star Wars - HarthDarth "Roxors!")

[personal profile] lorax 2011-02-26 10:37 pm (UTC)(link)
It's fine, tone is difficult to read on the net after all. Sorry for the misinterpretation.

it's potential users we need to attract, and that's a vastly different use case. And much more diverse.
Understandable. Thank you for the clarification.