Google interface.

Forum Moderators: open

Message Too Old, No Replies

Google interface.

Never saw this before...

The Contractor

11:23 pm on Jul 29, 2003 (gmt 0)

Just ran across a visit (that included the users search term) from this in the logs of a site I watch. You have to view it in IE and everything is backwards - checked with others to see if they saw the same thing....

[google.co.il...]

tedster

11:38 pm on Jul 29, 2003 (gmt 0)

It's Google's Israeli search engine - and Hebrew is read from right to left.

The Contractor

11:48 pm on Jul 29, 2003 (gmt 0)

Thanks for clearing that up :)

I showed several people that and they thought it was a prank.....

caine

11:52 pm on Jul 29, 2003 (gmt 0)

There are many languages in the world that one cannot read, but this is one of the oldest, i have to wonder how G or the plug-in programmers actually got this recognised from a binary translation point of view.

Tell me if i am wrong, but i always thought that all computer language is English. Obviously the text is UTP-8, but is english still the only logical langauge in which processors. Funny thing is that Indian (India) is producing most of the software in the world. But how the logical translation from a non latin-greek (similar characters - though not all is used) to a lanuguage that looks like an art form amazes me, just can't understand the character transformation and G's implementation of it, to the search audience - full marks for G - it keeps on trying to get better - obvious, shame there ain't another couple of engines that look at the world of improving the info search, and maybe making some money along the way.

Marcia

12:22 am on Jul 30, 2003 (gmt 0)

It's been quite a long time, but from what I remember it all gets compiled down to zeroes and ones with compiled languages, regardless of which it is. In the case of COBOL on mainframes it would compile as IBM Assembler and to go through you had to wade through enormous printed out storage dumps of hexadecimal notation to debug.

We actually had someone from Israel letting us know in Forum 3 that there was some sort of a problem with Google/Israel search a while back, so there must be something extra in dealing with the right to left language and different notation.

The Hebrew language and some Arabic languages read right to left rather than left to right; I've sometimes wondered whether that affects how they decide to design pages. I believe some Western designers put their navigation on the left and "Click Here" on the right side for conversion effectiveness because of the natural eye movement going left to right in reading. I wonder if others take the opposite approach to accommodate right to left reading.

killroy

12:39 am on Jul 30, 2003 (gmt 0)

The problem you're referring to related to teh byte ordering in the files. Basically should the html file read:

or:

>lmth<
>ydob<
>ydob/<
>lmth/<

And how to deal with each case.

Appearantly both are valid ways of encodign right to left wreiting, and google had some troubles outomatically recognising which one was used.

g1smd

11:09 pm on Jul 31, 2003 (gmt 0)

The content of the page, the stuff that visitors can read, can be done as right-to-left; however, I don't think that you can do that to the HTML tags themselves. I'm sure that will not work.

swizz

4:41 am on Aug 1, 2003 (gmt 0)

>lmth< ... LOL :-)

Shalom

- SwiZZ

shrirch

5:22 am on Aug 1, 2003 (gmt 0)

[w3.org...]

Defines the direction of the language.

Hebrew is right-to-left (rtl)

kaled

11:10 am on Aug 1, 2003 (gmt 0)

Right-to-left or left-to-right is purely a DISPLAY issue. Either the first char of a page is placed at the top-left or top-right. It is absolutely no big deal. On the other hand, right-to-left languages tend to be pictorial in origin (so far as writing is concerned) and therefore require a much larger alphabet. This is a much bigger problem. Character strings need to be encoded either as two-byte pairs (often called unicode) or variable-length chars (one or two bytes with the length determined by look-up table on the first byte or by a flag bit). And then there are issues of code pages - YUCK.

It would make sense to allow HTML tags to be written backwards, so that they could be read in English when displayed right-to-left. I cannot comment on whether this is part of the HTML standard, but writers of browser software would not have a problem implementing this.

Kaled.

cococure

8:53 pm on Aug 1, 2003 (gmt 0)

I noticed they have the scrollbar on the lefthand side as well. I didn't know you could do that.

caine

9:06 pm on Aug 1, 2003 (gmt 0)

thats been possibly for a while. JS. ASP. PHP. Flash. Active-X. many languages are capable of the left hand scroll bar, not just a client readable website, but also browser configurable code accompanying the site.