Bikeshedding is very common when building things in the open. You get used to it, and you learn which comments are contructive and which ones to ignore pretty quickly.
One thing I still don’t understand though, is this resistance to making things better: denying there is a problem, and trying to think of reasons why you should not improve things. This isn’t because “your design sucks” or “you should have done it this way”, those kind of comments can actually be useful to an extent. The kind of comments I’m talking about are the ones oblivous to the points made and those that just resist change for no good reason.
I bumped into it again on this article at OMGUbuntu! (the writer has good points). Sure this is OMGUbuntu! which can be considered the new Slashdot (comments) from time to time, but because this one was about the font install dialog in GNOME that was redesigned by me and Thomas Wood about two years ago, it reminded me of the same comments we got at the time.
I did a bit of quote mining. Now some of these could be taken with a tongue in cheek kind of way, but it wasn’t very clear. You can read the comments yourself too.
Um… Just drag the folder into Fontmatrix, confirm you want to install those 20 fonts, tick families to activate
I don’t know if you hunted those chocolate eggs, in my ~ the .fonts folder isn’t really far away. Just had to copy fonts there and update the cache.
how is putting fonts in folder called .fonts not intuitive? And setting up a preconfigured copy operation is very easy. You can even add this folder to your favourties on the left and just drag the files onto it…
On ubuntu just double click… Only power users have to install a lot of fonts at same time, and they’re smart enough to locate the .fonts folder!
A simple custom action in Thunar would solve this.
sudo apt-get install nautilus-actions
I always install fonts manually (both on Windows and Ubuntu) just because I like it.
kde has a font installer does what you need
You can install fonts via synaptic. There is a package there that has 500mb of fonts.
This article is really a non-issue unless you install twenty fonts at a time, in which case, opening a file manager is really nothing hard and much better than cluttering up the right-click menu.
I found the same kind of backlash (altough to a lesser extend) in my post about the first redesign.
there is (was) an easier way to install fonts in gnome just write “fonts://” in the address bar of nautilus and drag and drop the font you want to install
I’m confused. Why is putting something in /.fonts so hard? When I download a font, the archive manager asks me where I want to stick it. So I say, “Archive Manager, my good man, put it in the .fonts directory,” and it is done. And the next time I fire up whatever program, lo, there is the font.
Why is this, and what can we do about it?
We are kinda planning to unscrew Fontmatrix’s UI for things like importing (think Banshee progress indication in the sidebar), but Pierre is super busy these days, and I’m no programmer, so it’s put on hold for now.
A lot of people like to solve problems. Something that they understand and works for them in their own computer’s context.. that’s the kind of quotes I think you have there, like “this is my solution”.
When I respond to user bugreports I sometimes say “Ah yes, we can certainly design XX better. In the current version you can workaround this using YY then ZZ.” After telling about the workaround you mostly never hear from the user again, which is a shame—user feedback is needed for a better design.
As for the GNOME tool in quesstion, I don’t really use it — not because it fails on multiple opened fonts (see below), but because it’s not verbose enough for my purposes (and especially because it fails on “national” fonts that don’t have “typical” Unicode coverages).
There are different ways to handle multiple opened fonts via your tool, and one of them is to simply open them all in one window, while generating a list of each font file to be able to navigate between them. Then you can click “Install” to install them all. You can complicate that dialog by adding a checkbox for every font file to enable/disable installation, but that will most likely be as bad as eating a GNOME designer’s baby
Users are not supposed to know about ~/.fonts from the very beginning.
I fail to see how those comments you posted are strong “denial” or “resistance” signs.
There is a problem in the article you posted (multiple font installation) and those people have discovered a rather simple/sane solution for it (copy files into ~/.fonts) and explain it in the comments. This solution will fit them, they will remember it forever and it will be their optimal quick solution.
I think the core issue for you here is that not everyone is, thinks like or learned to think like a foss developer or user interface designer. You can’t do anything immediate about it as you can not “teach” the world in an instant.
A lot of people simply don’t understand that anyone (even designers who can’t code) is actually able to act and try to fix something on this desktop (just post a bug and let other devs join in or post a patch if you can). So while the problem could easily be solved with “select multiple font files in nautilus and install them using the context menu”, you’ll hardly find someone proposing this.
The “font install button” saves research time (”how to install ttf in linux”) for novice users and even saves those experienced “copy to ~/.fonts” gurus a nasty few clicks. It would be ignorant bliss to say this would be bad or “resist” it’s use.
I think the overall goal should be really to offer obvious and sane features in the GUI preventing long research times for novice users. That is called “being more intuitive” and afaik that is what GNOME3 aims to be.
This kind of resistance has previously been called “stop energy”. http://www.userland.com/whatIsStopEnergy
Sometimes negative design feedback is useful. I suppose it’s an “I know it when I see it” kind of thing. But if you trained a Markov filter to tell the difference between useful and useless feedback, the “Useful” bucket probably would contain things like this:
- “have you thought about”
- “how about””
- “instead of”
- “would make it harder to”
- “would make it slower to”
While the “Useless” bucket probably would contain:
- “average user”
- “devs”
- “intuitive”
- “just”
- “simply”
- “sudo”
- “Um…”
- “you people”
Totally agree with ulrik, there are so many bug reports in which the user marks the status as FIXED when he has found a solution, not realising that the problem should be fixed for everybody so there is no need to report a bug ever again.
“I have learned this hidden and arcane way, so anyone else could just learn that too” - never mind that even explaining how to type ~/.fonts could be a challenge to some users. Not to mention updating a cache…
I’m a long time geek and very well versed in the command line, but I don’t want to have to do that step either, I mean why would any sane person want to keep a suboptimal way to do anything. Moreover, why would a GEEK (or hacker) want to keep a suboptimal way to do things, that just boggles the mind.
http://www.ted.com/talks/kathryn_schulz_on_being_wrong.html
Let me try to explain the sentiment you’re running into.
1) It’s a response to a perceived attack on linux and the linux way. Someone says it’s bad that linux can’t do this this way, people are responding that there are a number of great ways linux can do it. Pick any one of them, whichever way your picky sensibilities prefer, and you’re good. So what if windows has this one particular way to do this one particular thing and you already happen to know about that way, we’re not just going to copy windows button for button.
2) Old school unix pieces, like files and processes, are easier to work around and fix. When a shiny UI with a fancy automatic backend starts doing something wrong, it’s a pain in the ass to fix. If the font UIs eventually turn into a super easy automatic service where you can use a GUI to add, remove, enable, and disable fonts, and all applications seamlessly update their font lists in real-time, and are served fonts out of a magic dynamic cache over dbus or some crap, and it has a glitch, the old-school unix guys just smack their foreheads and install slackware, all they wanted was dejavu or terminus anyway.
I’m not going to explicitly support this camp, just wanted to express what I thought it was about
I remember commenting or filing on a bug that had something to do with installing fonts a few years ago, and I’ve been thrilled by the addition of the Install button on the simple dialogue. I’m no designer, so it suits me very well.
Thanks for your continued work and effort
I agree that the stop energy can be pretty frustrating. Sometimes I think people just want to chime in and say “this would be nice, too” or “here’s how I do it”, but they don’t actually mean to divert the plan. Unfortunately, it can still add a lot of noise.
By the way, thanks for adding that feature to the font previewer. My fiance likes to play with new fonts a lot and she was really happy to find out how easy it was to download a font and then actually use it.
“[…] those that just resist change for no good reason”
I think there _is_ a good “default reason” for many people, which is “why put lots of work into making it better in the future, when it already works so-so right now?”.
As a long-time open source freak I have really slipped into this mind set of “I can change _anything_!”, and “I can actually make improvements which will benefit other users as well!”. Spelling error in Wikipedia? Edit -> fix it -> Save -> Look at the new and shiny article. Missing street on OpenStreetMap? Online Editor -> fix it -> Save -> wait for some days -> look at the new and shiny map. Bug in gedit plugin? fix it in ~/.gnome2/gedit/plugins/ -> use shiny new plugin. Bug in Nautilus? bugzilla -> open bug report -> wait for some months/years -> use shiny new Nautilus.
However, I think most average users are far away from this mindset, for several reasons:
- not everybody can program
- even if people resort to making bug reports instead of fixing things themselves, it usually takes a while until a developer has looked at the problem, even longer until it is fixed, and yet longer until the fix appears as update to their desktop (usually at least 6 months). Add a fair share of bug reports which are never fixed at all, and you get users who don’t feel like “I can change anything” at all.
So I think people don’t really deny that there’s a problem, they just take the mental shortcut to “let’s ignore problems which are basically unsurmountable anyway”.
Sadly, as the Linux desktop evolves in usability, the type of people who would read a Linux-related blog (power users) will be the ones talking to you about the change. Unless the person talking to you is a designer or a developer themselves, they might resort to these kinds of ‘why change it when I already knew how to do it the less obvious way??’ comments.
And the users who benefit the most from your improvements, new users, aren’t the kind to comment even on OMG! Ubuntu. So the real issue is the breadth of feedback you’re able to get. Sometimes you just have to go with what makes sense, not what’s popular.
I think GNOME 3 is a good opportunity to innovate and renovate things based on the best possible ideas for interaction with our computers and the information they hold. It’s better to try something new and tone it down later than to simply try growing an old idea.