Will GTK be used in Openmoko?

[Stroller]

On 16 May 2008, at 06:19, Shachar Shemesh wrote:
> ...
> Oh, so that's alright then, is it?
>
> Do remind me how many people live in China. Last time I checked,  
> traditional Chinese would not accept left to right input, nor  
> simple key presses, nor went below a sixth of the world's population.
>
> ... in Japan... Look at the phones around you and do tell me how  
> many of them have English only input. I don't know phones very  
> well, but for general typing, last time I checked the smallest  
> subset of Kanji was 120 characters long, and was only used by the  
> Government. People typing solely in Hiragana was not something  
> Japanese would look nicely upon. And that's Japanese, a language  
> that converted from writing top to bottom with columns going right  
> to left to writing left to right with lines going top to bottom. Do  
> keep in mind that the rest of east Asia has not done that  
> transformation, to the best of my knowledge.
> ...
> Israel has a much higher level of English literacy among the  
> general population than almost any of the Arab countries. You can  
> walk down ANY street of ANY city and ask for directions in English,  
> and it will be very rare to have to get a confused look and no  
> answer. When I work with my phone, I use an English interface,  
> because Hebrew technical language looks weird for me. However, when  
> I look at the vast majority of friends phones, they are on Hebrew  
> interface. You live in a non-English speaking country. I suggest  
> you do the same.
>
> Maybe a huge chunk of the world can GET BY on left to right, single  
> letter at a time, but it is far from true to say that that is what  
> a huge chunk of the world prefers.
> ...
> in the developer community. Then again, the NEO is not intended to  
> stay within the developer community.

Hi there,

I've cut your response - and edited the placement of only the last  
line - in order to focus on this one point. You seem to think that  
upside-down languages (if you'll pardon my informality) should be  
supported _now_ because Openmoko is planned _eventually_ to ship on  
mass-market phones.

I wrote in a previous message how sales estimates of the Freerunner -  
and FICs hopes for sales into the next several years - are likely a  
trade secret which are unlikely to be published openly on this list  
and which are really none of us end users' business. One doesn't,  
after all, ask one's neighbour what his income is. However it's my  
belief that FIC have a 5-year business plan for their phones, and I  
think that one probably shouldn't expect to see Openmoko-based phones  
in high street stores for at least a couple of years. I think there's  
a general expectation on this list is that "woah! we're gonna get the  
`developer version' next week, dude! they'll surely be in the shops  
way before Christmas!!"  (PS3 f0r teh w1n!!:!:! teh 36o 5uX0rZ!!)

I think it's more likely that the Freerunner - or any Openmoko-based  
phone - is likely to sell almost exclusively to existing Linux geeks  
for the first year of its production. I would imagine it would sell  
largely to speakers of European languages for some more years, and  
that FIC would be very glad to make significant inroads to the mobile  
phone markets of only those countries (I include Brazilians,  
Americans &c &c as speakers of what I consider broadly to be  
"European languages"). Sure, FIC'll be more glad to make inroads into  
markets with languages in other character sets, but that shouldn't  
blind us to what is economically feasible & expedient.

I would guess - from my experience working for a very average  
software development house - that you could employ 10 average visual  
basic programmers and not get half such good results as employing one  
Rasterman, who is surely a little cheaper. FIC is a new company and  
hasn't sold anything yet - in a business situation like this one  
watches one's costs and tries to get the best results one can without  
spending too much money. It's simply not an option to employ 10  
Rastermen right now and add all the features one might like - it's  
Rasterman's job right now to get out of the door the best, most  
functional mobile phone that he can, considering the constraints of  
his time and the likely customer base.

Magic does not happen overnight, and I think our enthusiasm for this  
product may be blinding many of us to the long-view - I'm sure, for  
instance, that Gnome is much different from the version I tried when  
Madriva was still Mandrake (perhaps in 2002??) and that many advances  
have been made. In a couple of years' time we might well see  
significant changes made to Openmoko's UI, but that's something to  
think about then - everything's different when you've sold a few  
thousand, or tens or hundreds of thousands of phones. FIC will have  
_income_ - that's REALLY significant!! FIC doesn't have income at the  
moment, and maybe in a couple of years time FIC'll tear everything  
down and start the software again from scratch (Openmoko-NG) to  
accommodate the language changes you desire. Right now, FIC's  
priority is to get hardware out the door, and to get working phones  
in peoples' hands.

Operating-system language support is obviously something that's  
important to you, but if you take a look at the economic realities, I  
think you'll agree that there's no reason for it to be a priority  
right now (sorry).

Stroller.