preventing fork of Firefox

6 views
Skip to first unread message

Dmitry Turin

unread,
Nov 21, 2008, 8:23:25 AM11/21/08
to wish...@lists.mozilla.org, dev...@lists.mozilla.org, dev-pl...@lists.mozilla.org
Hi, all,

We are intending to hire any company to implement new HTML-elements [1] in Firefox
(communication between browser and DBMS should take place via sending SQL and XML [2]).
We have the following questions:

1) Is our conclusion right, that [1] takes more than just writing code to get Firefox to have a new
feature, and it needs updating Gecko ?

2) If so (and we would not want to make fork of Firefox), how can we obtain agreement of developers,
that features to be in the main Firefox version ?


[1]
http://html60.euro.ru/site/html60/en/author/looker_eng.htm
http://html60.euro.ru/site/html60/en/author/tree_eng.htm
http://html60.euro.ru/site/html60/en/author/cube_eng.htm
http://html60.euro.ru/site/html60/en/author/screen_eng.htm

http://html60.euro.ru/site/html60/en/author/forxml_eng.htm
http://html60.euro.ru/site/html60/en/author/hidden_eng.htm
http://html60.euro.ru/site/html60/en/author/db_eng.htm

[2]
http://sql50.euro.ru/sql5.16.4.pdf

Dmitry (SQL50, HTML60)


Sergey Yanovich

unread,
Nov 21, 2008, 9:29:52 AM11/21/08
to
On 2008-11-21 15:23, Dmitry Turin wrote:
> We are intending to hire any company to implement new HTML-elements [1] in Firefox
> (communication between browser and DBMS should take place via sending SQL and XML [2]).
> We have the following questions:
>
> 1) Is our conclusion right, that [1] takes more than just writing code to get Firefox to have a new
> feature, and it needs updating Gecko ?

Not necessarily. The whole [1] thing can probably be implemented as an
extension. My company is working on a much less ambitious project with
similar functionality.

> 2) If so (and we would not want to make fork of Firefox), how can we obtain agreement of developers,
> that features to be in the main Firefox version ?

My experience as an independent Mozilla contributor says that this is
going to be tough to sell. Packaging [1] as an extension can mean
slightly more work compared to patching Gecko/libxul, but provides much
better chances to get massive field testing.

--
Sergey Yanovich

Boris Zbarsky

unread,
Nov 21, 2008, 10:38:56 AM11/21/08
to
Dmitry Turin wrote:
> 1) Is our conclusion right, that [1] takes more than just writing code to get Firefox to have a new
> feature, and it needs updating Gecko ?

Everthing in that list looks like it can be implemented in XBL or
XBL+XTF with no core gecko changes.

> 2) If so (and we would not want to make fork of Firefox), how can we obtain agreement of developers,
> that features to be in the main Firefox version ?

Convince them that they're useful to someone other than yourself?

-Boris

Dmitry Turin

unread,
Nov 24, 2008, 7:20:25 AM11/24/08
to dev-pl...@lists.mozilla.org
Hi, all.

> Everthing in that list looks like it can be implemented in XBL or XBL+XTF with no core gecko changes.

Thank you, Boris.

How XULRunner corresponds to XTF ?
What is the relation with these two terms ?


Dmitry (SQL50, HTML60)


Dmitry Turin

unread,
Nov 24, 2008, 7:23:01 AM11/24/08
to dev-pl...@lists.mozilla.org
Hi, Dev-planning.

And what else:

Ian Hickson <i...@hixie.ch> from publi...@w3.org said us, that
it will take more than just writing code to get Firefox to have a new feature.
I admit, he is wrong, but i'd like to listen your opinion - he is wrong ?


Dmitry (SQL50, HTML60)


Benjamin Smedberg

unread,
Nov 24, 2008, 9:06:56 AM11/24/08
to

There's a critical distinction: you can probably implement this as an
extension, and then ask users to install your extension. Doing this doesn't
require agreement from anyone else. If you want this functionality to be
part of Firefox (by default) you'd have to make a case that it is actually a
worthwhile thing to do.

In any case, you'd do well to start out as an extension to demonstrate the
value of whatever these new SQL+HTML proposals.

--BDS

Boris Zbarsky

unread,
Nov 24, 2008, 9:45:48 AM11/24/08
to
Dmitry Turin wrote:
> How XULRunner corresponds to XTF ?
> What is the relation with these two terms ?

I answered this in the .platform thread you started on this exact same
topic. Please stop cross-posting, and please do some googling before
asking questions like this, ok?

-Boris

Dmitry Turin

unread,
Nov 24, 2008, 12:00:53 PM11/24/08
to Boris Zbarsky, dev-pl...@lists.mozilla.org
Hi, Boris.

> I answered this in the .platform thread you started on this exact same
> topic. Please stop cross-posting, and please do some googling before
> asking questions like this, ok?

ok :)


Dmitry (SQL50, HTML60)


Daniel Veditz

unread,
Nov 25, 2008, 3:55:26 AM11/25/08
to
Dmitry Turin wrote:
> Ian Hickson <i...@hixie.ch> from publi...@w3.org said us, that
> it will take more than just writing code to get Firefox to have a new feature.
> I admit, he is wrong, but i'd like to listen your opinion - he is wrong ?

If you want your code in the shared Firefox repository then you have to
make a case that the benefit from the new feature (to firefox users as a
whole) outweighs the costs. Costs include increased memory use, the need
for additional testing, any loss of performance, and increased attack
surface for security vulnerabilities.

For example, you want to enhance SQL. Firefox already uses sqlite as a
SQL engine and it would be insane to have two. You might get what you
want if you convince the sqlite folks to implement your new SQL
features. If not, you'd have to convince Mozilla folks to replace sqlite
with your new engine which sets a performance bar you'd have to meet.

btw, HTML5 proposes some support for SQL, are you sure that doesn't do
what you want? It would be far more effective to work within a standards
structure the browser vendors are already cooperating in if that's possible.

Dmitry Turin

unread,
Nov 25, 2008, 6:29:07 AM11/25/08
to dev-pl...@lists.mozilla.org
Hi, Daniel.

> benefit from the new feature (to firefox users as a
> whole) outweighs the costs.

+
> the need for additional testing

It sounds like benefits Fortran before assembler.

> Costs include increased memory use
+
> any loss of performance

Recall about MS Windows, and even Linux now is not well.

> increased attack surface for security vulnerabilities

To not be in theoretical flood, by what mechanism in my case ? ??

> For example, you want to enhance SQL.

It is redundant for HTML6 - you can connect to DBMS via its branded API.

P.S.
SQL5 is proposed as transport protocal for universality - between all DBMS
and all client (by transfering SQL), between traditional result sets and
hierarchcal results sets (by transfering XML).

> Firefox already uses sqlite as a SQL engine

I'm not user of Firefox now - interest is only for our project.
So i can mess some features, correct me, if i will wrong.
Yesterday i spoke with some user of your browser, he said me about
html-element <template>, apparently you are talking about it ?
We compared, and <template> is very week for real automatization
instead of my <dialogue> ...

> and it would be insane to have two.

... so speach goes not about using concreate DBMS, but about _right way_ of
using.

> You might get what you
> want if you convince the sqlite folks to implement your new SQL
> features.

Not SQL features, but features of

Can you seggest, what procedures i have to execute to convince ? ??

P.S. the following

are really for SQL5, but it is quite other project.
So now speach goes only about 4 papers.

> If not, you'd have to convince Mozilla folks to replace sqlite
> with your new engine which sets a performance bar you'd have to meet.

Replacing engine by more performance (e.g. Postgres) - is right idea.
But i would leave this sub-task to future time.

> btw, HTML5 proposes some support for SQL, are you sure that doesn't do
> what you want?

As i know, it does not do - we discussed about it on publi...@w3.org .
Can you give link to what you suppose ? ??

P.P.S.
Daniel, do you participate in dev-pl...@lists.mozilla.org ?
Maybe we will go into this list ?
As Boris Zbarsky asked :)

Dmitry (SQL50, HTML60)


Daniel Veditz

unread,
Nov 25, 2008, 12:15:39 PM11/25/08
to
Dmitry Turin wrote:
> I'm not user of Firefox now - interest is only for our project.

That is probably a sign that your interests and the interests of Firefox
users may not be aligned.

> Yesterday i spoke with some user of your browser, he said me about
> html-element <template>, apparently you are talking about it ?

There is a XUL element, but I'm unaware of a proposed HTML <template>
https://developer.mozilla.org/en/XUL/template

>> btw, HTML5 proposes some support for SQL, are you sure that doesn't do
>> what you want?
>
> As i know, it does not do - we discussed about it on publi...@w3.org .
> Can you give link to what you suppose ? ??

http://www.whatwg.org/specs/web-apps/current-work/#databases

The What-wg work may be a super-set of the official HTML 5 working spec,
but browser vendors are working with both (but have not committed to
supporting all of it, it is a work in progress).

Google Gears is a browser addon that provides many features, including
database support. You may wish to check it out and see if it fits some
of your needs.

Dmitry Turin

unread,
Nov 26, 2008, 11:06:59 AM11/26/08
to dev-pl...@lists.mozilla.org
Hi, Daniel.

>> I'm not user of Firefox now - interest is only for our project.
> That is probably a sign that your interests and the interests of Firefox
> users may not be aligned.

You are wrong wholly: requirement to browser does not depend of
vendor of browser.


> https://developer.mozilla.org/en/XUL/template

Very much code. With very sophisticated parameters.
So it is unavailable for regular user.
Must be one html-element, user will not master more.


> http://www.whatwg.org/specs/web-apps/current-work/#databases

It is shit, because needs JS, unavailable for regular user.
I laugh, how people cannot step over border, between static html-elements
and interactive html-elements.

> Google Gears is a browser addon that provides many features, including
> database support.

http://code.google.com/apis/gears/api_database.html
Identical shit by identical reasons.


Dmitry (SQL50, HTML60)


John J. Barton

unread,
Nov 26, 2008, 11:26:05 AM11/26/08
to
Dmitry Turin wrote:
...
> You are wrong wholly:...
> It is shit, ...

> Identical shit by identical reasons.

Well *I'm* hoping you fork.

John.

Dmitry Turin

unread,
Nov 26, 2008, 11:37:17 AM11/26/08
to Benjamin Smedberg, dev-pl...@lists.mozilla.org
Reply all
Reply to author
Forward
0 new messages