[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Netscape goes GPL.
On Sat, 24 Jan 1998, Eli Marmor wrote:
> 7. During this discussion, someone compared Hebrew support for X to
> Titanic project. As almost all of you know, I already did it, in
> addition to dozens of other projects I did. It works perfectly,
> and all the UNIX vendors in Israel already purchased it (in
> addition to many software houses and large sites). He claimed
> that a full compatibily is hard to be achieved; Well, my product
> is fully (100%) compatible, in both source AND binary level, to
> the original X (11.5 AND 11.6) and Motif. And its BiDi is fully
> compatible to Microsoft (although visual order is supported as
> well). You can take a Motif application, just replace its libXm
> (by changing LD_LIBRARY_PATH), and have the full Hebrew features
> that this application has under the Hebrew Windows.
> By the way: this "someone" said: "but it can be done, the
> foundation is there". Well, the foundation is not there anymore.
> It is very sad, and I knew all the staff (I enjoyed a week with
> them).
I did not mean "The Foundation", but the technical codebase required to do
it. The one that one does not have to port anything made by M$ in any kind
of way. The one that allowed you to do it without first signing an NDA or
making an investment that you could avoid (in buying a source license for
example).
>
> 8. One claimed that freeing the sources of Netscape will ease the
> localization of it to foreign languages, including Hebrew. First
> of all, I don't know what is wrong with my Hebrew support for
> Netscape. But secondly, and much more important, is the following
> fact: There is an exact correlation between the "commerciality"
> of software, and its chances to be localized to Hebrew. There are
> thousands of commercial software packages, and thousands of free.
> Among the commercials, 200-300 have Hebrew support. Among the
> free? 3, maybe 4. I can explain the reasons of this amazing fact
> in hours, but I don't have the time for it. Just remember one
> thing: The chances of Netscape to have Hebrew support, are lower
> than ever.
>
> 9. One claimed that freeing the sources of Netscape will improve it.
> Unfortunately, the technical problem of Netscape is not lacking
> features. Moreover, I think Netscape is too rich in features. And
> IE too, of course. The problem of these two browsers is that
> instead of being just browsers, they became monsters, with
> hundreds (or even thousands) of small features. So both are
> unstable, crashing again and again, and have severe memory leaks.
> Freeing the sources will allow Netscape to have many more
> "exciting" features which nobody will need, but the stability will
> not be improved.
>
> 10.As a (claimed to be) expert in Hebrew localization, I must admit
> that Microsoft is doing very well. Although everybody has many
> complaints regarding to their Hebrew support, I still didn't see
> anybody (including me) who has a competing Hebrew. It is sad to
> say, but we are all, years behind Microsoft, in Hebrew
> localization, who set new standards in it (and don't laugh at me)
Hmm. I'm not laughing. I am thinking of my W3.11+Hebrew which never could
get around those two little buttons at the top of each window (the Hebrew
<-> English selector).
Anyway, until UNICODE or something like it takes over completely, there
will be problems all the time. The only UNIX package that handles more
than one language in the same docmuent reasonably well is LaTeX and its
relatives. I know NO package under M$ that does something even remotely
like it.
Seriously, it is not *that* hard to have a table of fonts in the browser
under UNIX, and to have the default font reloaded according to the MIME
Charset given in the URL header. If we will have the source for N then
that will be one of the first priorities.
I don't want to make myself look big, but I'd like to say that browsing
with any browser is a big problem for me. I switch 3-4 languages all the
time and the browser can't even switch 2. Do you know what a German page
looks like with "ae" an "ue" shown as dots all over the place ? Or the
French dichars encoded in the 'high' 8-bit codes, looking like <B7><65> ?!
If you think that M$ has got the localization problem cleared then you are
probably not so badly hit by this problem. International companies have
very much the same problem, and in Europe things are worst.