I Don't Want To Be A Nerd!

The blog of Nicholas Paul Sheppard
Archive for october 2012

Building a social network, one auto-generated message at a time

2012-10-28 by Nick S., tagged as social networks

Today I received an invitation to join ResearchGate, which I gather to be a kind of social network for scientists. I'd never previously heard of ResearchGate and, almost certainly, they'd never heard of me. I nonetheless warranted an invitation because I co-authored a number of papers with someone who had already enrolled.

I mostly reject automated invitations of this sort, in part because I resent web sites expanding their business by taking advantage of my relationship with a third party and in part because I like to think that my real friends would be bothered to write real e-mails. But my experience of LinkedIn is my greatest motivator.

At the time I received my LinkedIn invitation, I had no experience of such sites and it seemed worth a try. But I never found anything useful I could do with it, and I gradually realised that my LinkedIn page was a graveyard of ex-colleagues who had sent me connection invitations but with whom I no longer actually communicated (via LinkedIn or otherwise). I began to wonder if sending a LinkedIn invitation was a tacit declaration that "I will never talk to you again."

After a few years of this, I began replying to invitations with a personal e-mail explaining that I don't really use LinkedIn. In response, one of my would-be connections admitted that she didn't really use LinkedIn either, but she just felt compelled to click on the "Do you know?" buttons. I was already pretty sure that my own LinkedIn connections were a fraud, and my friend's message suggested to me that I'm not the only one. I've since deleted my LinkedIn profile, and I refuse all new invitations with an e-mail explaining that I don't use LinkedIn.

It still feels slightly rude to reject invitations, and perhaps LinkedIn members feel it would be rude to ignore the question "Do you know?" when they do, indeed, know that person. I wonder if we instead ought to feel rude for allowing Internet companies to exploit our relationships in order to build their customer bases, and to present false social networks built up by automated messaging and idle button-clicking?

Software interpretation of human communication and its discontents

2012-10-23 by Nick S., tagged as communication, user interfaces

I found myself using Google Mail today, having joined a company that uses it as its e-mail system. Aside from my typical frustration with the bloated, slow and browser-dependent interfaces sported by modern webmail programs, I was specifically annoyed that Google Mail, by default, hides part of the e-mail that I'm working on: namely, the signatures and quoted material. As a result, I found myself reading and sending e-mail messages without being sure how they appeared on the receiver's screen.

By coincidence, I also received an e-mail today from a friend apologising for the poor formatting of her previous e-mail. Apparently it looked fine in her e-mail client, but it was garbled by other people's clients (including mine). I recently received another e-mail with a signature containing two different e-mail addresses for the sender, one obviously wrong.

My experience with Google Mail illustrates why this might have happened: the e-mail clients of the victims in the above stories presented a view of an -mail that they deemed helpful, while other people's e-mail clients presented views that those clients deemed helpful.

The views disagreed. Instead of presenting the "true" content of the e-mail, the e-mail clients involved have presented their own interpretation of it. (By "true", I mean the universally-agreed encoding of e-mail messages, being plain text or at least HTML.) And, in doing so, the e-mail clients foiled human communication.

Regarding my particular complaint, I suppose that Google Mail's developers think they are being helpful by automatically eliminating "extraneous" information like quoted messages and signatures. As I said in my previous post, I'm all for eliminating useless distractions from user interfaces. But if signatures and quotes really are useless distractions, why include them in an e-mail in the first place?

I'm part of what seems to be a dwindling minority of people who adhere to the custom of selectively quoting the e-mails to which we reply. Once upon a time, Internet users would have been appalled at the wholesale quoting of earlier e-mails that seems to be current practice. If you're not going to use it, we thought, delete it and save the space.

Now, Internet bandwidth isn't as precious as it used to be, and one might say that only hoary old nerds would cling to byte-pinching practices developed in the days of 2400 baud modems. But I nonetheless think that selective quoting serves a more human purpose: it foregrounds what is important to the communication, and eliminates what is not.

Interventionist features like Google Mail's seem to me to at once encourage lazy communicators to fill their e-mails full of junk in the expectation that their software will correct it for them, and to frustrate careful communicators by making them fight against their software in order to send the message that they want to send. The winner is poor communication.

Who's for clutter and distraction?

2012-10-13 by Nick S., tagged as user interfaces

A few weeks ago, The Register featured an article entitled Information is the UI in Windows 8, says design guru. I read the article 3-4 times and still have no idea what the eponymous design guru (Shane Morris) was on about, but the comments on the article reminded me of one or two battles I've fought and lost over user interfaces. I was reminded of this again while trying to disable a particularly annoying feature of LibreOffice the other night.

I'm a dyed-in-the-wool minimalist. I detest desktops full of icons; I fight constant battles against Windows programs that want to add themselves to prominent places on the "Start" menu, task bar and desktop; and I despise Gnome and KDE for emulating Windows. (For the record, I prefer Fluxbox.)

There is obviously a certain amount of personal preference here, but the comments on The Register's article make it clear that I'm not the only one. A few commenters mention the infamous <blink> HTML tag. When it was first developed, it presumably seemed like a good way of emphasising text. However, people rapidly discovered that blinking text distracted readers from the rest of the web page.

Google's home page is, I think, a legendary piece of web design. It replaced the complicated and distracting interfaces of Altavista, HotBot, et al., with an interface that gets right to the point of what people want from a search engine: a "search" box. Yet web designers -- apparently unaware of Google's success -- continue to stuff their sites full of links, images, Flash, Javascript and all the rest.

I don't think I've ever heard anyone say that the way to make a good user interface is to add as many buttons, sidebars, animations and other doodads as possible. So why do well-known web sites like Bigpond and NineMSN seem to do exactly this?

The obvious answer is "feature-itis", the process by which software gains features that seem good in themselves but whose over-abundance as a whole detracts from the comprehensibility and usability of the software. It's easy to say that some software or a web site should provide such-and-such a feature, and usually easy enough for a software developer to make the feature happen. But the proponent of a feature is unlikely to admit that it isn't that important and can be relegated to a second-level menu.

One of the pieces of information that I used to resolve my problem with LibreOffice illustrates this. A user of similar mind to me complained that the section title tooltip displayed by LibreOffice when scrolling through a document is "distracting and annoying". In response, Roman Eisele asserts that "some people will find this tooltip useful". So, on the strength of some unidentified people who might conceivably like this feature, Roman suggests resolving the original user's complaint with a feature that allows anti-tooltip users to disable the scrolling tooltip without disabling all tooltips (which is what I had to do in order to disable a feature that, for me, makes LibreOffice virtually unusable.)

I suppose there must be people who like interfaces that I find cluttered: plenty of computers have them. According to the November 2012 issue of APC Magazine (p. 57), Microsoft decided to remove the "Start" menu from Windows 8 in part because users "were pinning their favourite apps to the taskbar instead," which I consider to make for a very cluttered taskbar. By way of pleasing these users, the Interface Formerly Known As Metro seems to pile icons onto the screen in exactly the way that I hate most. Mind you, most of APC's writers seem to have grave doubts about it, too. I wonder if Microsoft has also checked that those users weren't just pinning icons to the taskbar because the "Start" menu itself is a bastion of uncontrollable clutter?