Dot, squiggle, plop
My first response to the idea of a dot-sex web domain was: "Well, that's a rip-off". And when I heard about dot-tv and dot-euro I wasn't the only one to say: "Here we go again..." And I bet most of us were equally cynical about dot-asia. I mean, most people in Asia don't actually use the Roman alphabet and so (I sort of assumed …
"But now it's been done (and is in beta) the next step is going to be, surely, to decide whether we can open up the rest of programming technology? Was that a worm I saw?"
If it was, it then it was made of straw.
Why does ensuring that consumers can enjoy a website and enter URLs in their native encoding mean that we need to start writing code in it?
Despite many attempts over the years to make writing code 'simple' or even 'easier' learning to code is all about learning to write IN ANOTHER LANGUAGE. Indeed, I've always suspected that non-English speakers manage to avoid some of the common early mistakes that English speakers make when thinking in English and writing code with English prepositions ;-)
And having coded on a Mac for some time, I've worked with a number of different nationalities who've written native comments in any number of languages...
Because there's already 260 of them:
http://www.google.co.uk/search?as_q=&hl=en&client=firefox-a&rls=org.mozilla%3Aen-GB%3Aofficial&hs=ZPa&num=10&btnG=Google+Search&as_epq=&as_oq=&as_eq=&lr=&as_ft=i&as_filetype=&as_qdr=all&as_occt=any&as_dt=i&as_sitesearch=.asia&as_rights=&safe=off
Who would get first dibs of dot-seman?
Hat, coat, door.
As you mention, most programming languages are written in American; I think the word English should be avoided unless it is. There's no reason a programming language can't be written in English, or in any other language / script; in fact I'm sure there are some out there, I think BBC Basic spells colour correctly at least. America tends to dominate because of its population and economy. I'm sure if someone wrote a PHP type language in Hindu it would spread pretty fast. The problem would be getting a world-wide user base, especially in the rich western countries.
Errr, the first page are all domain registrars and the second page says the rest are duplicates so there are no real sites yet.
I think they're just registering at the moment and won't go live until next year.
There was some attempt to include expanded character sets that ended up with a lot of exploits using the expanded capability in the couple of browsers that tried to support them and I also seem to recall a lot of arsewits downing said browsers for even making the attempt even though it was ICANN's intent that it should happen at some point so really what the problem is you can't even try to experimentally support kanji or cyrillic without a bunch of static and morons bitching about things they don't understand in the slightest.
.tv is the TLD for Tuvalu. It's not some special TLD for TV channels. While I'm here I can also tell you that .im is for the Isle of Man, and not some special TLD for instant messaging, I could go on :) Personally I am all for the new TLDs I look forward to my own .xxx domain.
One of my personal soap-boxes, this. How hard could it be for browsers to recognise <centre> and "colour" and treat them the same as <center> and "color"? It would take pretty much no effort at all to allow English words (as well as Americaneze ones) in languages like HTML. So why doesn't it happen?
Of course there are such things, and have been for a long time, but for obvious reasons our English-speaking editor has never heard of them. They tend to remain local to their "linguistic area", and even there are just a small niche for elementary programing education, or end-user scripting. Major programing projects need sophisticated tools with a larger base of both tools and experienced developers, which these local languages normally lack.
I'm aware of one case where non-English based language was used for big things: a lot of public administration software in Finland used to be written in "FAS", which according to some examples I have seen (punched cards reused as children's hobby material!) used keywords based on Finnish. [By the way, does anyone else reading this know of a good reference for the language? There is very little about it on the net]
There's nothing new here. I remember seeing French Algol 60 compilers back in the late 60s. Its simple to do - just replace the compiler's token list and parser rules with the appropriate tokens and rules for the language of your choice:
debut
si a[i] = b alors allez L1;
fin
..and these days we have Unicode, so its easy to extend to Cyrillic, Katakana and even idogrammic languages. Of course, a Mancunian won't be able to read a Cyrillic Java program unless he can read Russian, but that's a different problem. Its called "learning foreign languages at school" though it may not be too difficult to write dialect translators for a programming language.
I don't see the same problem applying to web pages, because most people won't be interested in looking at pages in languages and alphabets they can't read, so not being able to understand http://squiggle is a non-issue. Besides, its what Babelfisf Mk II is for. Its not exactly difficult to load several tag sets into a web browser so that it can correctly display pages coded in other languages. Granted there is an issue with right-to-left text, but as browsetrs can already deal with Arabic, this is scarcely a show stopper.
Last but not least, sorting out these problems with web browsers and graphical interfaces will give the browser and OS writers something to do. It will stop them getting bored and noodling around adding useless "features" to their programs.
C# / visual studio already DO support arbitrary charactersets... the following works:
double π = Math.Atan(1) * 4;
MessageBox.Show(π.ToString());
"There's nothing new here. I remember seeing French Algol 60 compilers back in the late 60s. Its simple to do - just replace the compiler's token list and parser rules with the appropriate tokens and rules for the language of your choice:"
Er. Obviously having made sure that any variable/object names in source code don't end up becoming reserved words by the compiler!
Lets hope we don't try to make the compiler's thousands-separator locale-specific. Ouch.
The embarrassment of not having a .uk top-level domain causes needless work for pretty much everyone in .co.uk Britain. If we can't have .uk, can we at least have .gb or .eng?
"Er. Obviously having made sure that any variable/object names in source code don't end up becoming reserved words by the compiler!"
In principle, this cannot happen in Algol 60. The spec requires the reserved words are special symbols that cannot collide with variables. How this is done is implementation-specific. Some systems required the keywords in quotes. See "stropping" in Wikipedia: http://en.wikipedia.org/wiki/Stropping
---------------------------------------
One of my personal soap-boxes, this. How hard could it be for browsers to recognise <centre> and "colour" and treat them the same as <center> and "color"? It would take pretty much no effort at all to allow English words (as well as Americaneze ones) in languages like HTML. So why doesn't it happen?
---------------------------------------
What would be the point? It's just adding additional bloat to the browsers and making them even less small form-factor friendly (more disk/memory intensive)... ok, in you're example it's only a teeeeeny bit. Let's also add <mitte> <zentrum> etc etc etc....
I'm not overly bothered as to which language it's based on AS LONG AS IT'S STANDARDISED! (hmmm used an 's' there, I must be English... can't even standardise standardise).
@Waggers
You are joking, right? 'American' is the lingua franca of the computer world, get used to it or find another profession.
How hard would it be to make browsers understand 'colour' as well as 'color'? Not too hard. But what about kleur, couleur, farbe, χρώμα, 色, 顏色, 색깔, цвет
And that's just a few languages. Now what about understanding alternatives for 'Header'? H1, H2, H3? The H is for header. In French then this should be T1, T2, T3, etc.
And so on for all the other tags and attributes. Then account for languages that don't read left-right-down.
If you REALLY want to write a browser engine to do all that then Gecko is open source, so you could start modifying that.
Good luck.
I'm pretty sure if you know what you're doing you could write a Firefox snap-in that would allow the browser to decode standard English into Americanized dialect. The problem lies with (spit spit..) Internet Explorer.. It still has 60% of the market share and I'd say writing a similar snap in would be damn near impossible.. And Microsoft ain't going to do it because there's no money in it and anyway, to them everything's spelled correctly already.
I think the easier way to do this is to write a coding program that would allow you to create "alias's" in your written code that the would allow you to use the language set of your choice to write your code and the program switches the code automatically, you hit a button it writes the same code in standard HTML and hit it again and it codes it back using your alias list. It would be a pain to administer at first, but I'm pretty sure that it wouldn't take long before the open source community compiled alias lists to suit your language, making web design and so on a cinch for non-American English speakers. I'll bet there's some cash to be earned if any of you felt like writing this?
By the way, I noticed on Wikipedia (don't ask, I can't remember why I was looking at this subject!) That there is a move to merge the word knickers into the word panties.. Those of you in the English (as opposed to the American) speaking world.. We should do something to stop this travesty.. There's a link from the web page to discuss the changes.. http://en.wikipedia.org/wiki/Knickers That's just such an ugly word!! Please help keep your language sacred!
There have been many examples of programing languages with kewords in different languages; French Algol is only one. The issue of keyword homonyms due to unicode keyspace was also extensively discussed in the unicode world years ago.
But the fact is that it's no more alien for a non-English speaker to use "<font>" or "for (i=0; i<x; i++);" than it is for an English speaker to use "Allegro" or "Da Capo a la Fine." Or eat camembert or sushi. Or even say "OK".
People world wide understand all of these "technical" and "alien" terms. What's the big deal?
I remember attending a seminar about DNS back in 2001 where I was told that internationalisation, iirc Unicode characters, was just around the corner. BIND9 has supported it for some time.
Wah lau! What are these "Singaporean" characters of which you write? English and Bahasa Melayu both use the Roman alphabet, Mandarin is written in Simplified Chinese, while Tamil uses... Tamil.
Sheesh. What a straw-man argument! Other than a few words off of the local Greek-American restaurant's menu, I don't speak a lick of Greek... but I can program in APL *just fine.*
If your example had *any* merit, programmers would need to "Speak" assembly... how many languages is LEAX -1,X or ASLB a real word in, anyway?
[[ OK, for those few of you that recognize the above as Moto 6809 assembly, you may also note that there _is_ an SEX instruction. As much as I wish it worked on the wifey, it doesn't... but it will Sign EXtend the B accumulator across A. ;-) ]]
But it would make phishing so much easier. It wouldn't exactly be difficult to trick people into entering their login credentials to the domain "еЬаy.com" thinking they were getting "ebay.com"
It is not .euro for the european domain, it is .eu, (but I think it should be .€ just to piss of the septics)
By the way, BBC Basic isn't the only British programming language, so was BCPL the first "curly bracket" programming language. This was used by a certain Dennis Ritchie to create the B complier and from that came C -> C++ -> C# -> Java -> PHP, etc. We mustn't forget HTML of course - thank you Sir Tim - even if it can't spell colour.
Pascal was created by Niklaus Wirth - A Swiss - and the keywords aren't French, German, Italian or Romanche! This whole discussion is a red herring, really. The language of the compiler is irrelevant to most programmers, n'est pas?
So many worms... sounds like a Phishing expedition.
While .tv was originall createdas the ccTLD for Tuvalu, that country sold it to somebody else for some insane amount of money (relative to their GDP. Something like $8000 per head of population - it's a small island). So, while it wasn't created as the Television TLD, it has effectively become that since the country that owned it,sold it.
Sign up, sign up for The Register's weekly mobile & wireless newsletter - click here
