Dodo Birds Of Linux History

I was sorting out a pile of old CDs this morning and found my first linux install discs, in spiffy looking cardboard:

They were researched, greatly wished for, highly anticipated, and a huge letdown. Continue reading

Consumer Fake Out: Your Nightlights Are Lying To You

This past week I picked up a couple of cheap LED nightlights at Marden’s ($1.99 for a two-pack) and had to take them apart, of course, as I am a self-respecting geek, and because I was at one time mildy obsessed with running LEDs off line power. I guess that hasn’t quite worn off.

This all has led me to two conclusions:

  1. American consumers are suckers
  2. Chinese are either idiotic electrical engineers or brilliant social engineers, or both

Actually, both of those are not entirely deserved, nor totally fair, but read on for the full tear-down. Continue reading

The Joys Of Deleting

There is an odd social pressure to collect stuff. This may be due to our tendency to buy the latest and greatest in good consumer fashion, and it may be due to our belief that more information is good information and you can never have too much (not true). It is likely a combination of both and a variety of other factors which I have overlooked.

In any case, we both consume and store more and more information every day and most technology is concerned with getting it to you faster and cheaper, whether it be your smartphone, your media server, or Gmail. I think that is wrong. Continue reading

Rotary Encoders Are Mostly Awesome

Rotary encoders are awesome because they function like knobs – a staple of user interface which needs no explanation – but  electrically they generate discrete button presses, like pushing the Up or Down tuning button. They are reduced to mostly awesome by some designers who apparently don’t know how to use them correctly.

Case in point: the Kenwood VR405 surround receiver I recently picked up from the side of the road.

It’s a cheap and basic surround reciever, but it works (other than missing the volume knob), it was free, and it has a cool-looking rotary encoder on the front panel. Rotary encoders are like the bling of tactile input devices. Buttons are nice and all, but things that spin are just so much cooler. They also happen to work really well for linear controls like volume and tuning, for which they have been used for decades and still should be more than they are. (I’m looking at you, designers of 1980-s electronics-with-no-knobs and of infernal-car-stereos-with-button-densities-approaching-infinity.)

However, Kenwood got it completely wrong on this model: the input selector is rotary and the tuning/everything else “Multi Control” buttons are Up/Down buttons. Unlike rotary selector switches of yore this input selector wheel has no detent to tell your finger it has reached the next setting. This makes changing inputs quickly is very easy but stopping on the right one is very difficult. And tuning? Click, click, click, click, click, as fast as your finger can push the button you can change to the next frequency. Tiring. Why on earth didn’t they use the up/down buttons for the input selection and the rotary for tuning, like, you know, old-school tuners used to be? Some of the pricier models in the V/VR series were slightly better because they had a rotary encoder for both the Input and Multi controls, but Kenwood could have saved a couple pennies and given users a less frustrating UI at the same time by just using rotary for the Multi Control and buttons for the Input Selector across the whole line.

Heavy Weather

I don’t think I’ve ever seen that orange icon, until yesterday, at which point it was novel. Now, if I look carefully, I think I can see R2D2 and C3PO wandering across dunes. I never thought I’d look forward to 90° weather in Maine.

Gravely wins, Ice loses.

No match for the big metal after a two-day thaw. We love our machine.

Dichotomy Of A Real-Life Maker Online

I love the maker movement. The wide spread community of people who love building and modifying things for fun & profit, practical and just-because, resonates with my natural inclinations. If it breaks, fix it. If you need something, build it. Modify stuff until it works for you and makes your life better in some way.

This is the great thing about the internet – we can now all share the myriad of things we’ve built, hacked, or dreamed of, and be inspired to create things of usefulness & beauty. The only bugger is that I get so busy being inspired that I don’t get anything done. I could (and have, may times) spent hours reading hackaday or Instructables or the Make: blog. It’s awesome. And colossally time-sucking. In which lies the rub; for in my quest of the cool and useful inspiration I find myself with increasingly little time to actually build the things already on my list.

The principle holds for most of the internet, really. It’s an incredible resource, a wealth of knowledge on almost any subject imaginable and then some that aren’t. The magnitude of information at the fingertips is both exciting and mind-boggling, because I love information. I love knowing and learning about things, I love being able to search the world for opinion, experience, and data at a moments notice.

At the same time, I have trouble managing large amounts information. If I’m not careful, the ever-branching search for whatever bit I was looking for (or just happened across) inflates exponentially. That is when I find myself, two hours later, with thirty more browser tabs and enough reading material for another week of evenings – and still with my list of projects I’d like to actually, physically do. Never mind the stack of books and last month’s Popular Mechanics waiting patiently on the end table.

Ah, dichotomy. A Maker doesn’t just consume, a maker creates, taking ideas and raw stuff and turning it into something better, something useful. How can I create when my I leave no space for it? All the thinking, reading and book-larnin’ doesn’t cause completed things to magically appear. I have to actually get out and do something. So if you see less of me around the internet, it’s because I’m living more in real life. And building stuff, too.

Change Of Face(book)

Over past Facebook changes I have largely, roundly, and imaginatively derided complainers. I have a very long, somewhat impolite, and happily unpublished blog post to that end somewhere, regarding the last facebook restructuring, and I still have very little patience with people who whine, cry, and generally carry on obnoxiously about every single thing Facebook changes. No good reasons, no quantitative analysis, or in depth discussion – mostly just a pity-party of complainers.

I know, most people may not be good at analyzing and comparing things, and/or don’t have the time – but they still find the time to complain without a second thought. The least they could do is be respectful about it, and (*gasp*) say something about why they don’t like it. Face it: posting things like “it STINKS!!! i hate facebook now it sucks and its different and what were you smoking?” does not inspire any respect at all and is mostly un-helpful.

So in the interest of my own opinions on whiners, I will not whine about the changes. I will attempt to compare the current UI iteration thoughtfully to past versions and provide insight into why most of you hate facebook so much right now, or at least did last week. And I apologize for the length of this post, but it seems I can’t do things like this by halves. To my knowledge, this is the most thorough analysis of Facebook’s interface changes ever published.

Anyway, here’s my nut-shelled opinion: I think the layout took a couple steps forward and a whole bunch of baby steps backward. The changes have been made in the name of simplicity and ease of use. It is definitely more simple. However, it is visually less structured (i.e., more cluttered), does not always work the way I want or expect it too, and does not provide me with as much control or value. If it does not do what I want it too, ease of use tanks as well. This is a case where simpler is not necessarily better.

Goodness

Filtering based on friend lists
This is really nice – it’s the one major feature I wanted in the old facebook that wasn’t there.

Layout
The layout *is* simpler. For the most part, the underlying concept of the new facebook is pretty good, with exceptions noted below.

Badness
In the Operational Unintuitions Category
(Or, assumptions about user behavior which might not be quite right)

The post box operation is nebulous
It looks the same everywhere, but can behave very differently depending what page it’s on. It is completely unclear that posting on your own wall is different than posting on someone else’s, which has been very aptly demonstrated by my newly-facebookified mother, who replied to multiple wall posts by setting her status. The concept of a “status” is completely hidden unless you already know how it works.

  • This could be easily remedied – for starters, change the post button text to say “Set Status” when on your own profile page, unless posting a special type of content.

(See also the note on post format – either change the format so it works better, or split statuses back out to a separate entity. And the “What’s on your mind?” line is a bit hokey, but I can’t think of anything better than “Post stuff” or “What do you want to share today?”)

The Recent activity thing is confusing
Labeling activity on the wall as “recent” serves no useful purpose for two reasons:
1) I can figure out if yesterday is recent on my own, thank you, based on the post’s timestamp
2) The “Recent Activity” boxes are all through the wall, even if the activity is no longer recent. This makes the term “Recent” totally useless.

This is/was done (I think) because it is easier to filter programmatically and visually groups wall posts and informational stuff separately. However, it would still be better to use the “Today”, “Yesterday”, “Some other day” headings, grouping the posts and other activity underneath. The smart date headings were great because they are very simple, require no interpretation, and (most important!) give the user control over what they think is “recent” enough to be interesting. I honestly don’t care if it’s recent, I just care if it’s interesting to me and generally when it happened.

The single “delete” button is not intuitive
When I click it, I do not expect a particular friend to stop showing up on my news feed, I expect to just get rid of (or de-rate) that particular post or post type. The friend-based behavior is redundant – news feed supposedly already rates somewhat based on friend activity/interaction. This is an example where simplification is completely irrelevant – it’s simpler, sure, but it a) isn’t very useful and b) doesn’t work the way users expect.

Wall post filtering is not designed from a user’s perspective
This one is sticky. The wall has been restructured for ease of use by everyone but the person who owns the profile. The defaults should be to show user+friends, with option to show only friends or only user. The current settings emphasize broadcasting the activity of the user, rather than letting the user choose how they want their wall to work and be used.

Wall posts no longer have a “Reply” link
This is a UI dilemma – it’s a duplicate feature. Yes, you can just as easily post on a friends wall by clicking their name or on the “Wall-to-Wall” link, but if I’m replying, I look for a “reply” link. It’s not intuitive to click one of the other links. It would be amazingly intuitive to click “reply” and have a post box drop down (or pop up), like a comment, only it would post directly to the friend’s wall instead of commenting. This would be a killer feature.

The new feed is a lot of information in a huge, daunting list
It’s very hard to get useful information out of so much data. The old feed algorithm provided a much better balance and control over what showed up – now I have almost no useful control. See also the note above about wall post activity/date headings.

I no longer have control of post types in the News Feed
Before, there were preferences with sliders to customize the type of content that came across my feed. Now the only way to do that is manually, with application filters. Not good.

And now in the Visual Miscues section…

Post format is visually muddled
Previously, the name of the poster was in a sort of header and the message was separate. The current run-on format is harder to read because it is not visually well-defined. It works great for statuses, but nobody writes wall posts like statuses. Heck, most people don’t write statuses like statuses. For wall posts, a better balance between space and readability would be good – even just putting a break between the name and post.

The new feed is harder to read – no visual formatting!
The old news feed formatting was much easier to read – the “Today” “Yesterday” etc. headings broke down the news feed into smaller, more easily parsable chunks. (The abbreviated & more compact format helped, too.) Coupled with the more mashed post formatting, it’s much harder to read the news feed.

Miscellaneous UI quirks

  • The rounded images look icky and ill-defined. Keep the sharp, square, refined look.
  • Can we please put pokes with other notifications or something? Burying them in the middle of the right-hand sidebar has always bugged me, it makes no sense.
  • Group invites & friend requests should go back over on the right hand side with events, pokes, etc. They get lost visually at the top, even if it makes sense grammatically.
  • Highlights are pretty useless. It looks messy and is distracting. I want the relevant stuff I like to show up in my feed, not over there. Reformatting them could help, or put some sort of visual break between the news feed and highlights.
  • People You May Know should be hideable. It’s useless to me – I want the option of not seeing it, like I had in a couple versions ago with apps on the profile page.

My Big Solution:
Put all invites, events, pokes, and notifications on the right-hand sidebar. Group as necessary for ease of use. Limit the amount of space notifications can take up, or make it say “You have 23 notifications” and the link unzips it.

So there’s my take. Hope it’s useful.

How Not To Design Stuff: Dellusions Of Power


The University of Southern Maine employs the use of a large number of Dell Optiplex GX280 computers in labs and classrooms. They are modern computers, pretty fast, quiet, not real pretty, and the power button of these computers looks pretty much like any other typical power button on the planet: it’s round, it has that funky little circle-with-a-line-symbol-of-power-thingy, and it glows green. However, one unfortunate design flaw has caused lost time and headaches.

How, you ask, can the design of a power button result in confusion and lost work? Visual feedback.

Fact: The button glows green when the computer is on.
Fact: People expect something on the computer to glow green when the computer is on.
However: In order to tell that the button is glowing, one must look almost straight at it from the front. If you are a little to the side – which is the normal arrangement for the Labs – you can’t see the light.

Result: If the computer is positioned anywhere but in front of the user, the little green glow is invisible unless one leans over and peers at the button.

One would not think this to be a terrible problem, a dinky little light doesn’t make that much difference. After all, you can also tell if the computer is on by looking at the monitor.

That logic is conveniently defeated by two different circumstances.

1) Some classroom computers are attached solely to a projector – if the projector isn’t on (or is blanked), there is no feedback from the screen.
2) In the John Mitchell Center CAD lab, all the monitors are on a separate power circuit and can be turned off at the professor’s whim.

We are back to using the power button glow as the sole indicator of computer status. This is a dangerous prospect to entertain if the glow cannot be easily seen – and did I mention that pushing the power button when the computer is on immediately shuts it down?

(Cue scary theme music.)

You push the button. Nothing seems to happen for a couple seconds. You push the power button again. The computer you just turned on shuts off again. You wait longer this time, and are confused. You push the power button again, firmly, and wait longer and something finally shows up on the screen. You can now do useful work before class starts.

Then, the monitors are shut off during lecture and you forget the computer is on – and there are no visible glowy lights to tell you either way. The lecture is over and you push the power button to start the machine up. It shuts off. And, since the computers have DeepFreeze (or equivalent) installed, anything that you saved to the internal drive is wiped clean at boot.

It sounds ridiculous, but I’ve seen people lose valuable work because of that. Extra frustration, problems, and lost work – all because the power button doesn’t provide a good (i.e., visible!) visual cue.