57184 - warehouse.com - menus only support document.all

Status: RESOLVED FIXED

(Tech Evangelism Graveyard :: English US, defect, P3)

Description

[Mike Pinkerton (not reading bugmail)]

Mac branch build from 10/17/00, but also bad on win32, 10/11/00

http://www2.warehouse.com/default.asp?home=mac

lays out very poorly. All the content is at the bottom, below the left sidebar.
Sending to attinasi as a starting place.

Comment 1

[Mike Pinkerton (not reading bugmail)]

nominating for rtm, bad layout makes us look pathetic.

Comment 2

[Chris Petersen]

The text that is displayed is used in several Layer elements.

Comment 3

[Chris Petersen]

Sample of one of the Layer elements:

<Layer CLASS=stdMENU id ="stdMENU1" z-index="1" bgcolor="#f5f5f5" top="65"
left="140" visibility="hidden" onMouseOver="stdshow('stdMENU1')"
onmouseOut="stdhide('stdMENU1')" >  </layer>

Comment 4

[Mike Pinkerton (not reading bugmail)]

cc'ing hyatt, as i'm sure he'll want to watch the joys of this bug play out...

Comment 5

[Marc Attinasi]

It is the layers. I'll attach a local copy of the page with the layers removed
(via CSS) and it looks reasonably good.

My guess is that the best we can do here is evangelize the webmaster to get them
to update their page to use standard constructs instead of layers.

There are also some strance css rules that use a tag name of 'all' that do not
match anything in Mozilla, but apparently do in Nav? Anyway, changing the
'all.stdNewSub' selector to '*.newSubSelector' fixes the menu coloration problems.

Removing RTM and correctness keywords, adding evangwanted
 and cc'ing ekrock.

Comment 6

[Marc Attinasi]

Attachment: Local copy of page with LAYERS hidden via CSS; looks fine

Comment 7

[Mike Pinkerton (not reading bugmail)]

adding 4xp, which is what ekrock said to do in the all-hands

Comment 8

[ekrock's old account (dead)]

No, sorry if I was unclear. What I said was that fixing 4xp bugs for quirks will 
be useful post-RTM. LAYER/ILAYER/document.layers[] have all been eliminated from 
Gecko forever, so they don't count as 4xp. See 
http://sites.netscape.net/ekrock.standards.html . Removing 4xp keyword.

--> Evangelism component. Removing evangwanted keyword since the evangelists 
apparently want a bug *either* assigned to Evangelism component (when it's a 
pure evangelism-need bug) *or* marked with evangwanted (when it's a bug report 
against the product that's legit for other reasons but still needs some 
evangelism).

Comment 9

[David Hyatt]

I believe we should revisit the decision to eliminate layers.  At the time 
this decision was made, XBL didn't exist.  XBL lowers the cost in time/effort 
required to implement them.
 

Comment 10

[ekrock's old account (dead)]

No. Dead snake. This decision was finalized in April '99. Resources that would 
be squandered on a futile attempt to provide bug-compatible emulation of the 
legendarily buggy behavior of LAYER/ILAYER/document.layers[] would be better 
applied to stability, performance, and memory optimization work over the next 
6-8 months that would be of value to all projects that will use Gecko. David, if 
you think this is the best use of your considerable talents, I can think of 
better ones! ;->

Comment 11

[David Hyatt]

I didn't say *I* would do it. :) Someone on the outside was wanting to give it a 
 try.

Comment 12

[Mike Pinkerton (not reading bugmail)]

this is a problem with all of the warehouse.com sites. changing url.

re-adding 4xp because "A bug is a Product Parity bug if it occurs on Mozilla
builds, but does not occur using the latest release of Netscape Communicator 4.x
or competitor products."

Regardless of the actual problem, it looks fine in previous products and our
competitor, but looks like crap in our product. Users don't care why or who's
fault it is, they just want to use the web.

Comment 13

[ekrock's old account (dead)]

Removing 4xp keyword again because this is a misuse of it. If you're 
concerned about tracking lack of backward compatibility due to lack of 
support for the LAYER tag, tag the Summary [LAYER]. That is the convention we 
use in Bugzilla for this specific purpose. Marking 4xp on Layer-specific bugs 
adds no information, blurs the intended meaning of that keyword, and wastes the 
time of everyone who does searches in the future trying to find 4xp bugs that 
are still considered candidates for fixing.

The 4xp keyword is intended as a tool to track product parity issues with 4.x 
*that the relevant module owners would still contemplate fixing in the future* 
and have not already decided not to fix. It's not a tool for tracking dead 
snakes. 4xp is not a way of tracking backward incompatibilities as some kind of 
a scorecard of total global incompatibilities; it's a tool for tracking backward 
incompatibilities that the relevant module owner believes are candidates for 
fixing in a future release. It's a tool for participant in mozilla.org to search 
for backward incompatibility bugs that are candidates for fixing, categorize 
them, and evaluate them and prioritize them.

This bug report is absolutely the wrong place to reopen the long-since-closed 
debate about whether Gecko should support LAYER/ILAYER/document.layers[]. If you 
want to do that, please open a newsgroup thread on n.p.m.layout which is the 
correct place. But that entire decision was debated at length over a period of 
months in the public newsgroups, the module owners based on that input decided 
that LAYER would never be supported, and the decision was publicly announced and 
finalized in April 1999 and endorsed by Jeffrey Zeldman of WSP among others. The 
decision has been publicly documented and communicated to the developer 
community which is planning accordingly.

Comment 14

[Mike Pinkerton (not reading bugmail)]

adding 4xp keyword, per the def on the keyword page. if you disagree with that
page, file a bug with mozilla.org

Comment 15

[ekrock's old account (dead)]

*** Bug 57330 has been marked as a duplicate of this bug. ***

Comment 16

[Blake Ross]

-> [e-mail] for my evangelism bugs.

removing the now-depreciated evangelism-related keywords.

setting platform to All.

Comment 17

[Boris Zbarsky [:bzbarsky]]

*** Bug 58480 has been marked as a duplicate of this bug. ***

Comment 18

[Boris Zbarsky [:bzbarsky]]

*** Bug 61797 has been marked as a duplicate of this bug. ***

Comment 19

[shao_lo]

php.net and msdn.microsoft.com have similar problems

Comment 20

[shao_lo]

The layers on php.net render properly now :)

Comment 21

[Blake Ross]

Reassigning evangelism bugs to [e-mail].

Comment 22

[Bob Clary [:bc] (inactive)]

All Evangelism Bugs are now in the Product Tech Evangelism. See bug 86997 for
details.

Comment 23

[Doron Rosenberg (IBM)]

they now handle us as if we were ie, doving us document.all foo. 

Comment 24

[Zach Lipton [:zach]]

Mass reassign of all tech-evangelism us general bugs assigned to bc to 
doron except bc's P1 bugs. You may search for this mass reassign (it is 
305 bugs) by searching for the keyword 'greeneggsandham'

Comment 25

[Frankie]

As Doron mentioned a year ago, the good news is that Warehouse no longer uses
the bad old LAYER tag.

The bad news is that their javascripts don't browser sniff at all, and assume
everyone uses MSIE.

Possible contact addresses are [e-mail] and
http://www3.warehouse.com/customerservice/default.asp?sel=contact&Subsel=feed

Comment 26

[Bob Clary [:bc] (inactive)]

tech evang june 2003 reorg

Comment 27

[Bill Daly]

Feature sniffing has been added to check document.getElementById before
document.all.  Site works fine now.

Change status to Resolved?

Comment 28

[Bob Clary [:bc] (inactive)]

Yep, thanks again Bill. We actually are better supported that IE. :-) -> fixed.