274256 - Site Evangelism letter needs to be revised

Status: RESOLVED INCOMPLETE

(Tech Evangelism Graveyard :: Other, defect)

Description

[Prognathous]

> Your site CUSTOMIZE: insert URL here does not support the Mozilla browser or
other browsers based upon the Gecko Layout engine such as Netscape 6, Netscape
7, AOL for Mac OS X, CompuServe 7, Galeon, and Camino(TM).

Firefox is becoming very common and should be top spot on this list.

> ftp://ftp.ecma.ch/ecma-st/Ecma-262.pdf

ftp.ecma.ch no longer exists. Try ftp://ftp.ecma-international.org/

> http://devedge.netscape.com

Netscape Devedge is gone. It should be hosted at mozilla.org soon.

Same problem here:

> http://devedge.netscape.com/viewsource/2001/updating-dhtml-web-pages/
...
> http://devedge.netscape.com/viewsource/2002/browser-detection/
...
> http://devedge.netscape.com/viewsource/2002/venkman/01/

In addition, we should add some text that explains how why we (as users) refuse
to use Internet Explorer due to its security issues. It doesn't matter that we
prefer Gecko-based browsers for other reasons as well, what matters is that
security issues are a "killer reason" to stay away from IE, a reason that can't
be easily dismissed by site owners (User: "Do you guarantee to send someone to
clean viruses and spyware off my computer if I use IE?").

A couple of related links:

"We are providing these instructions due to the recent disclosure of severe
security exploits in Internet Explorer and the recommendation to switch to an
alternative browser, such as Firefox, by the United States Computer Emergency
Readiness Team (US-CERT), a division of the Department of Homeland Security."

http://www.mozilla.org/products/firefox/switch.html ->
http://www.kb.cert.org/vuls/id/713878

"Is Firefox more secure than Internet Explorer?

    Yes, Firefox and all other Mozilla-based products are more secure. Why? Here
is a list of the most important reasons:

        * It is not integrated with Windows, which helps prevent viruses and
hackers from 
          causing damage if they somehow manage to compromise Firefox.
        * There is no support for VBScript and ActiveX, two technologies which
are the reasons 
          for many IE security holes.
        * No spyware/adware software can automatically install in Firefox just
by visiting a 
          web site.
        * Firefox doesn't use Microsoft's Java VM, which has a history of more
flaws than 
          other Java VMs.
        * You have complete control over cookies."

http://www.mozilla.org/support/firefox/faq#mozvsie

Prog.

Comment 1

[Prognathous]

Attachment: Revised Evangelism letter - v1

Attaching a slightly revised version of the Evangelism letter, with the
following changes:

1. Fixed a few broken links (ecma.ch and DevEdge).
2. Added Firefox and the Mozilla Application Suite to the list of Gecko-based
browsers.
3. Dropped CompuServe 7 (it doesn't appear to be available anymore).
4. Replaced Netscape 6 with Netscape 8.

Any other suggestions?

Prog.

Comment 2

[Prognathous]

Comment on attachment 187773 [details]
Revised Evangelism letter - v1

Bob, your comments?

Prog.

Comment 3

[Bob Clary [:bc] (inactive)]

(In reply to comment #2)

Sorry, I haven't responded sooner. Overall it looks like a good incremental update. 

> support the Firefox browser or other browsers based upon the Gecko Layout engine 
> such as Netscape 7, Netscape 8, Mozilla Application Suite, AOL for Mac OS X, 
> Galeon, and Camino(TM).

how about?

> support the Firefox browser or other browsers based upon the Gecko Layout 
> engine such as Mozilla Application Suite, Camino(TM), Netscape 7, Netscape 8, 
> AOL for Mac OS X and Galeon.

Too bad developer.mozilla.org isn't quite final yet, but we can reedit when the
temp url goes away. Do you think any of my documents on bclary.com about site qa
and testing are worth adding?

Comment 4

[Bob Clary [:bc] (inactive)]

developer.mozilla.org is live now.

Comment 5

[Prognathous]

(In reply to comment #3)
> how about?
> 
> > support the Firefox browser or other browsers based upon the Gecko Layout 
> > engine such as Mozilla Application Suite, Camino(TM), Netscape 7, Netscape 8, 
> > AOL for Mac OS X and Galeon.

It does make more sense. I'll update this in the next revision.

> Do you think any of my documents on bclary.com about site qa
> and testing are worth adding?

Many of them look useful, but since I'm not a webmaster myself, I'd rather have
someone more knowledgeable in the field decide which ones should be included.

(In reply to comment #4)
> developer.mozilla.org is live now.

It's live, but the DevEdge migration hasn't been completed yet. Here's the
current status of pages linked in the letter:

- Netscape Developer Web Site - 
New URL: http://developer.mozilla.org/en/docs/DevEdge

- Updating DHTML Web Pages - 
Not ready yet. Future URL will probably be:
http://developer.mozilla.org/en/docs/Updating_DHTML_Web_Pages_for_Next_Generation_Browsers

- Browser Detection and Cross Browser Support -
New URL:
http://developer.mozilla.org/en/docs/Browser_Detection_and_Cross_Browser_Support

- Reference Sidebars for Web Developers -
Migration started, no alternative URL yet.

- JavaScript Debugger for Mozilla, Netscape 7 -
This page hasn't been migrated and it doesn't seem like anyone is working on it.

The status of other pages is available here:
http://developer.mozilla.org/en/docs/DevEdge:Priority_Content

Prog.

Comment 6

[Gérard Talbot]

Proposed suggestions
--------------------

1-
The document
Using Web Standards in your Web Pages
http://developer.mozilla.org/en/docs/Using_Web_Standards_in_your_Web_Pages
should be mentioned in all letters.

2-
Several Mozilla-friendly or Firefox-friendly websites have actually discussion forum for web developers which, I think, should be mentioned in each respective letters:

Chinese: http://forums.mozine.org/ (all forums)

Czech: 
  category web development: http://forum.czilla.cz/index.php?c=3

Dutch: http://www.mozbrowser.nl/forum/ (all forums)

French: 
  For webmasters: http://www.geckozone.org/forum/index.php?c=4
  Web development: http://www.geckozone.org/forum/viewforum.php?f=8

German: http://www.firefox-browser.de/forum/ (all forums)

Hebreu:
http://www.mozilla.org.il/board/ (all forums)
(I'm abs. sure there is a forum for web developer in that page)

Norwegian: http://forum.firefox.no/ (all forums)

Polish: http://mozillapl.org/forum/  (listing all forums)

Spanish: 
  Web development category: http://www.mozillaes.org/foros/index.php?c=2
  Tech Evangelism: http://www.mozillaes.org/foros/viewforum.php?f=3
  Web development: http://www.mozillaes.org/foros/viewforum.php?f=13

Discussion forum in native languages may give web authors a more "live" assistance and more flexibility.

There are also web programming (eg javascript and html) discussion forums available in other languages.

3-
Personally, I would definitely and positively remove the link to the ECMAScript 262 3rd Edition: 99% of all problems related with webpages not rendering as expected in Mozilla-based browsers have nothing to do with the javascript language per se. 
99% of all problems experienced by webpages, website or users are related to invalid markup code, incorrect nesting, css parsing errors, use of document.all, IE-specific code, non-standard code, bloated code, etc.
In my opinion, sending web authors to visit the page
http://www.ecma-international.org/publications/standards/Ecma-262.htm
is an awkward reference to give.

Comment 7

[Gérard Talbot]

(In reply to comment #0)

> > http://devedge.netscape.com
> 
> Netscape Devedge is gone. It should be hosted at mozilla.org soon.
> 
> Same problem here:
> 
> > http://devedge.netscape.com/viewsource/2001/updating-dhtml-web-pages/
> ...
> > http://devedge.netscape.com/viewsource/2002/browser-detection/
> ...
> > http://devedge.netscape.com/viewsource/2002/venkman/01/
> 

Right now, Devedge is not mentioned anywhere in the template letters; only Mozilla Developer Center is mentioned.


> In addition, we should add some text that explains how why we (as users) refuse
> to use Internet Explorer due to its security issues. It doesn't matter that we
> prefer Gecko-based browsers for other reasons as well, what matters is that
> security issues are a "killer reason" to stay away from IE, a reason that can't
> be easily dismissed by site owners (User: "Do you guarantee to send someone to
> clean viruses and spyware off my computer if I use IE?").

I think we should keep the letter as short, straightforward and direct to the point as possible and stay away from issues which could be long/hazardous to develop. Remember that the template letters are for web authors, webmasters whose webpages give problems to Gecko-based browsers. How about just mentioning that, so far, Firefox has been downloaded 300 millions times and maybe a link to the "marketshare graph"
http://ff.asbjorn.it/pages/msgraph.php
We could then safely say, I guess, that together with Safari, Opera, Gecko-based browsers and other non-IE browser users represent 25% (or so) of the browser market.

> A couple of related links:
> 
> "We are providing these instructions due to the recent disclosure of severe
> security exploits in Internet Explorer and the recommendation to switch to an
> alternative browser, such as Firefox, by the United States Computer Emergency
> Readiness Team (US-CERT), a division of the Department of Homeland Security."
> 
> http://www.mozilla.org/products/firefox/switch.html ->
> http://www.kb.cert.org/vuls/id/713878
> 
> "Is Firefox more secure than Internet Explorer?
> 
>     Yes, Firefox and all other Mozilla-based products are more secure. Why? Here
> is a list of the most important reasons:
> 
>         * It is not integrated with Windows, which helps prevent viruses and
> hackers from 
>           causing damage if they somehow manage to compromise Firefox.
>         * There is no support for VBScript and ActiveX, two technologies which
> are the reasons 
>           for many IE security holes.
>         * No spyware/adware software can automatically install in Firefox just
> by visiting a 
>           web site.
>         * Firefox doesn't use Microsoft's Java VM, which has a history of more
> flaws than 
>           other Java VMs.

"after December, 2007, Microsoft will no longer support or provide a Java implementation with any of its products. Microsoft has already discontinued shipping Java technology in its Internet Explorer web browser with the latest versions of its operating systems."
http://java.sun.com/j2se/1.4.2/docs/guide/deployment/deployment-guide/upgrade-guide/index.html

"Microsoft will stop supporting the Microsoft Java Virtual Machine (MS VM) by the end of 2007."
http://java.sun.com/upgrade/

>         * You have complete control over cookies."
> 
> http://www.mozilla.org/support/firefox/faq#mozvsie
> 
> Prog.

If we address all these security-related matters/issues and IE-versus-Firefox comparison issues, then we would divert the letter from the main issue: their webpages don't work for Gecko-based browsers. I believe we should be very careful not to annoy people with advocacy; I believe we should focus and stick to technical support, useful, helpful assistance on W3C web standards.

Comment 8

[Gérard Talbot]

> Chinese: http://forums.mozine.org/ (all forums)

It is rather

Chinese: http://forums.mozine.cn/ (all forums)

Comment 9

[Gérard Talbot]

3 additional proposals
----------------------

1- Transliteration of browser names. I am now completely convinced we should stop transliterating browser names because even all people in forum assistance/support newsgroups in different languages (listed in comment #6) do not refer to Firefox or Seamonkey or Camino or Venkman or Internet Explorer, etc... in transliterated forms. It does not make a lot of sense either; transliteration is useful when adequate pronounciation is needed. Same thing with brand names, company names (Microsoft, Kodak, IBM, Hewlett-Packard, Canon, Ford, etc) in different countries: these names are not transliterated.

2- Based on rendering engine name. I also believe we should drop reference to rendering engine names as this does not help one bit web authors having problems with their webpages which do not work well or do not work as expected in
Gecko-based browsers. Trident, Presto, Gecko, Tasman, etc.: it's not important, not helpful, not useful to refer to these names.

3- Venkman reference. I think we should drop, should stop referring to Venkman as a recommendable debugger tool and we should replace it with Firebug. Firebug is most likely what web authors, webmasters would and could benefit from the most and most of the time. Venkman debugger is more of a professional tool for complex and advanced Javascript debugging. Well above 90% of the problems experienced by web authors in webpages which do not work (render, page layout) well or do not work as expected in Gecko-based browsers (or other acid2-compliant browsers like Opera 9, Safari 2+, Icab 3, Konqueror 3) are not related to ECMAScript but to W3C DOM coding (document.all, using id to access DOM nodes) or to markup coding/CSS coding (misnested elements, markup validation, CSS validation, invalid code, etc.)

Comment 10

[Felix Miata]

(In reply to comment #9)
> 2- Based on rendering engine name. I also believe we should drop reference to
> rendering engine names as this does not help one bit web authors having
> problems with their webpages which do not work well or do not work as expected in
> Gecko-based browsers. Trident, Presto, Gecko, Tasman, etc.: it's not important,
> not helpful, not useful to refer to these names.
 
Unless I'm misunderstanding your meaning, surely you jest. Better to strip Firefox, Camino, Flock, SeaMonkey, etc. from UA or evang discussion than Gecko. http://geckoisgecko.org/

Comment 11

[Gérard Talbot]

Referring to Gecko does not help by itself the web authors, webmasters who will receive such tech evangelism email letter. And they will be receiving that email letter because their webpages have problems (rendering, layout, overlapping, content not loading, etc.). 

How to upgrade their webpages, what to do, where to get help, assistance, how to correct problems so that *_all modern, web standards compliant, mainstream browsers_* render their webpages accordingly, etc.., that is what they will want+need to know, that is what the object of the letter is about, that is what their visitors would demand.

We can keep and we will keep Gecko in that email letter... but we must explain (clarify a bit more) that such word/name just identifies the rendering (or layout) engine of some browsers (Firefox, Seamonkey, Camino, Galeon, Epiphany, K-meleon).

------

Here's a major error I just found in the current English version:

"When you detect browsers embedded with the Gecko rendering engine, it's simplest for you to sniff for 'Gecko' rather than a specific vendor name or full user agent string."

But detecting the string "Gecko" is user agent string detection anyway (or browser sniffing) and is not reliable, not trustworthy. Object/feature support detection is widely known and recognized as the best, most reliable, most adequate and manageable way to write DHTML (javascript-driven) webpages. The given recommendation above goes against what other MDC web standards and MDC web development articles recommend.

Comment 12

[Gérard Talbot]

This comment is primarly for Prognathous and Shoshannah Forbes

I wish to have the "OK", explicit permission to stop transliterating 
- company names: Mozilla 
- browser names: Firefox, Internet Explorer, Safari, Opera, K-meleon
for bug 360302

There are several reasons in favor of not transliterating such names.

---------

I wish we could all reach unanimity on removing reference to ECMAScript from letters for reasons given in comment #6.


Also,
{
When you detect browsers embedded with the Gecko rendering engine, it's simplest for you to sniff for "Gecko" rather than a specific vendor name or full user agent string.
} < taken from http://www.mozilla.org/projects/tech-evangelism/site/tech-letter-en.html >
should be removed from all letters. There is now a wide and undisputable large consensus that object support detection and feature support detection is the best, most reliable, most forward-compatible, most relevant way to write cross-browser DHTML code.

Comment 13

[Chris Lawson (gone)]

INCOMPLETE due to lack of activity since the end of 2009.

If someone is willing to investigate the issues raised in this bug to determine whether they still exist, *and* work with the site in question to fix any existing issues, please feel free to re-open and assign to yourself.

Sorry for the bugspam; filter on "NO MORE PRE-2010 TE BUGS" to remove.