Memo to mobile handset vendors: what about MY smug, self-important delusions of entitlement and consumer empowerment?
Somewhere along the way, I became a relatively heavy user of mobile text messages. The geek factor probably plays a heavy role in this, as does the need to furtively communicate short thoughts while occupied at work. A girlfriend — and some friends — with a proclivity for the epistolary probably doesn’t hurt either.

Even so, I’ve never had a fancy mobile handset with a built-in QWERTY keyboard except for work-related purposes. In various jobs I’ve held, I’ve had a T-Mobile Sidekick II, a Blackberry, etc. For the most part, I’ve had to contend with banging them out on fairly conventional low-end flip phones; the kind that come as a free upgrade on a contract renewal with most major carriers.
Despite the awkward word entry scheme and seeming user interface limitations, I’m actually fairly comfortable with it and have gotten decently efficient at it. My gripe is not with the fundamental paradigm. I know that if I really wanted to type a lot of words into a mobile device, I should just get a PDA or a sophisticated data/voice portable of the variety mentioned above.
Here’s what I’d like to see in the SMS interface implementations in common consumer-grade handsets, in my ideal world:
- I employ proper capitalisation, punctuation, spelling, grammar and symbols in my messages. The interface should not be designed to discourage that. I want easy access to hyphens, semicolons, percentage signs and ampersands, not to navigate through fifteen menus to introduce them and derail my train of text-thought.
- If you’re going to make a good predictive dictionary (”T9″/”T9Word”), make a good predictive dictionary. I cannot count the number of things I have to switch to cumbersome manual entry for. Why aren’t most proper nouns, including names of famous people, places, or things, in the dictionary? Countries like Kazakhstan, people like Mobutu Sese Seko or Kim Jong-Il or Jean-Paul Sartre, or trade names like Pyrex? Solid-state memory gets cheaper and cheaper by the day, folks.
- I do not like how entering a wrong letter that forks off the predictive tree leaves me having to re-enter the entire word as opposed to merely correcting the mishap in the ending, etc. It seems that if I am trying to spell “something” and accidentally enter “somethhg,” I cannot delete three characters and recover the intended ending — even if it is one that would otherwise be a reachable leaf in the predictive tree.
- Why does the interface fail to make the entirely reasonable assumption that the next character within a word boundary following a period (punctuation symbol: .) ought also to be capitalised? By default, mixed-capital entry mode forces the outcome to “She shells sea shells. by the sea shore.”
- As a technology professional and a business enthusiast, I employ a lot of acronyms. But you don’t have to live in an acronym-ridden day-to-day world of banality to use them. I don’t expect the dictionary to know acronyms. But please don’t try to squash or mangle them into something other than what they are actually intended to be based on some erroneous assumption that normal people don’t use acronyms. Yes they do.
- This latest phone I have requires the navigation of far too many menus for the simple purpose of sending a message.
- Do we really need a 160 character payload limit in the SMS protocol specification? Maybe it’s just that the methodologies of writing for these presidium.org people have slowly rubbed off on me, since they pay me by the word, but there is very little of communicative significance or value that I can say in 160 characters or less.
- The predictor needs to be a lot better at learning my entry habits, and knowing which word I tend to use most of the time when entering sequences that map to multiple permutations. 95% of the time that I enter 7666, I mean “soon,” not “room.”
- The predictor needs to meticulously assimilate every new word I manually enter into its dictionary. It should automatically know when I have switched out of predictive and into manual entry mode to compensate for its ignorance and take careful note of what I am proceeding to type. Sometimes it does this, sometimes it doesn’t.
- The lexicon of the predictive dictionary should be cool, hip, trendy, with-it and modernity-affirming, and come prebuilt with words like “blog.”
That is all. k00l? thx 4 listenin, k… u 2 bye.
January 7th, 2008 at 4:26 pm
1. Um, yes. I don’t know how common this layout is, but I have a relatively low-end (”I-got-this-free-when-I-began-my-contract”) phone as well, and as far as text messages go, it groups everything that is not a letter or 0-9 all together in one list that I have to scroll through linearly. It’s not so much a problem that all symbols and punctuation signs are heaped together, but that they’re strung together like beads, and heaven forbid if I want a particular kind of bracket at the end of the list, oh nooooes.
Is there a layout for text messaging on any phones that are *not* of the PDA variety that set up these symbols on a grid? As in, something I could scroll through by rows and columns and reach my odd brackets more quickly?
4. My phone does assume this. Typing “happy time. jenqui” renders “happy time. Jenqui.” ¼ brownie point for Motorola.
8. I agree, but I do admit enjoying my phone’s occasional stupidity and/or creativity in that regard. I try to type “scene” and my phone suggests “Rafael.”
0. Is there a reason my phone does not admit to the existence of the caret ( ^ )?
January 7th, 2008 at 4:31 pm
1. From what I remember of your predictive input layout, it is something wholly foreign to what I’ve used — which is what is termed “T9Word.”
2. I have a table where I can scroll through by rows and columns to reach symbols relatively easily.
3. My old phone assumed proper capitalisation, too. I’ve downgraded and regressed, clearly.
4. The stupidity can be funny, but gets old fast. If the phone knows that I change “room” to “soon” almost all the time, it should present “soon” as an option for me first.
5. My phone does not support an upper carret either. It may not be part of the character set / encoding described in the SMS spec.
January 7th, 2008 at 6:30 pm
P.S.
I still can’t teach my phone “Jenqui.”
I do not think it is very Kazakh-friendly.